6a68ecb4d4b6faafee46a64366ceffbaacda2d02
543 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Jakob Wennberg
|
7240bfe7f3 |
fix(mcp): supplier-invoice-from-inbox resolves FX through the shared resolver, with the cache and an override (#2173)
* fix(mcp): supplier-invoice-from-inbox resolves FX through the shared resolver, with the cache and an override MCP feedback seq 299742: eight USD supplier invoices staged from the inbox in one batch, three got a Riksbanken rate and five came back exchange_rate: null / exchange_rate_source "lookup_failed", reproducibly, for ordinary weekdays in April-August. None of the five could be approved (the executor refuses with SI_FX_RATE_MISSING; it never books 0 SEK), and the tool offered no way to supply the rate. Cause: the tool called fetchExchangeRate without the supabase client, so neither the shared exchange_rates read-through cache nor the last-cached-observation fallback was reachable. Riksbanken's IP limiter answers 429 after about five requests in a burst (a weekend date costs two: exact-date 204, then the 7-day range), sends no Retry-After header, and asks for ~54 s, which the 5 s retry cap cannot honour. The pass/fail split was request ordering, nothing about the dates. - Resolve through resolveSupplierInvoiceExchangeRate with the client, the same resolver the commit executor and the v1/web write paths use, so the staging preview and the commit agree and the cache is consulted and warmed. - New input exchange_rate_override (SEK per 1 unit of invoice currency), trusted verbatim like the web form and v1; validated positive and finite, refused as implausible past the resolver's bound, rejected on a SEK invoice. Source is echoed as "supplied". - When the lookup still fails, the preview carries exchange_rate_hint saying approval will refuse and naming the override that unblocks it. tools/list: +1 property (~25 tokens), ledger line added in payload-size.bench.test.ts; ceiling unchanged. The retry cap is left as is: waiting a minute inside a tool call or the sync cron's fan-out is a design call, and the cached fallback now covers the common case. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013yw62FMXGSzo6icFDiBwP3 * docs(decisions): FX override and retry cap on the inbox supplier-invoice tool Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013yw62FMXGSzo6icFDiBwP3 --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
8265b5d166 |
feat(invoices): disclose invoice-register coverage gaps + net-amount search (#2122)
* feat(invoices): disclose invoice-register coverage gaps + amount search After a SIE migration or verifikat backfill, customer invoices exist only as journal entries: the invoice list, kundreskontran, /api/invoices, v1 invoices.list, and MCP list_invoices all looked complete while silently omitting everything before the register's first invoice (user report: two invoiced fees nearly re-invoiced as "uninvoiced"). - lib/invoices/invoice-register-coverage.ts: coverage boundary = earliest register invoice; flags posted non-invoice-engine AR verifikat (1510/1513) before it. AR-keyed, not source_type='import'-keyed, so manual/API backfills are caught too. - Invoice list page: one attn line disclosing the boundary (sv+en). - Kundreskontra: register_coverage in the report payload, rendered in the summary card and as an explanation under "Ej avstamd". - /api/invoices GET: invoice_register_coverage in the response. - v1 invoices.list: meta.coverage + registry pitfall documenting it. - MCP gnubok_list_invoices: invoice_register_coverage + coverage_note on the first page, pointing agents at gnubok_query_journal. - Search: lib/invoices/invoice-search.ts matches net (subtotal) and gross amounts with sv-SE formatting, alongside number/customer matching; a known net amount like 14 000 now finds the 17 500 kr row. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VcW5BU6mU1vNbWpkMKHbHF * fix(invoices): harden register-coverage probe, period-gate reconciliation note, regen api skill Skeptic + CI findings folded into one pass: - Coverage probe: a failed AR lookup now degrades to UNKNOWN (NO_INVOICE_REGISTER_COVERAGE), never to a confident "complete". - Probe driven from journal_entries (company-indexed) with the AR line condition as an inner embed, instead of the lines-table-with-embed-filters shape that lateral-scans every tenant (lib/bookkeeping/entry-lines.ts). - DEBIT-only 1510/1513 lines; excludes every invoice-engine source type (invoice_created, invoice_paid, invoice_cash_payment, credit_note, reminder_fee, rot_rut_payout, storno, correction): an advance payment crediting 1510 or a re-dated rattelse of an engine entry no longer flags. - covers_from ignores drafts so a backdated draft cannot move the boundary. - Kundreskontra "Ej avstamd" explanation is now gated on pre-register AR debits existing IN the reconciled period (new ARReconciliationResult.pre_register_ar_in_period): prior-period migration history cannot explain this period's difference and must not excuse a real felbokning. Wording no longer says "snarare an felbokning". - MCP coverage_note states the earliest register invoice date rather than claiming the register "covers" from it. - Amount search compares magnitudes so credit notes (negative totals) are findable; "-17500" parses; null amounts never match "0". - skills/accounted-api regenerated from the registry (apiskill:check). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VcW5BU6mU1vNbWpkMKHbHF * chore(api-skill): regenerate accounted-api skill after merging origin/main Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VcW5BU6mU1vNbWpkMKHbHF * fix(invoices): round-2 review fixes for register-coverage disclosure - covers_from now anchors on real invoices only (document_type='invoice', non-draft): proformas/delivery notes cannot move the boundary. - INVOICE_ENGINE_SOURCE_TYPES exported + a test scans the engine writers (invoice-entries, reminder-fee, rot-rut, storno-service) so a future source_type cannot silently become false pre-register evidence. - Kundreskontra guidance names both 1510 and 1513. - MCP gnubok_list_invoices outputSchema declares invoice_register_coverage and coverage_note. - v1 reports.ar-ledger documents data.register_coverage; invoices.list example made internally consistent; api skill regenerated. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VcW5BU6mU1vNbWpkMKHbHF * fix(mcp): keep gnubok_list_invoices outputSchema minimal to hold the tools/list token budget The expanded schema from the round-2 review pushed tools/list to 61 726 tokens against the held 61 600 ceiling (payload-size.bench.test.ts). The ceiling is policy, not a baseline to bump: the description already tells agents to read invoice_register_coverage/coverage_note, and paginatedSchema has no additionalProperties:false, so the fields stay schema-valid. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VcW5BU6mU1vNbWpkMKHbHF --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> |
||
|
Jakob Wennberg
|
67e878fb23 |
fix(mcp): reject unparseable voucher lines and allocation kinds; already-booked and unlinkable say why; N:1 reconcile groups survive staging (#2171)
* fix(mcp): reject unparseable voucher lines and allocation kinds; already-booked and unlinkable say why; N:1 reconcile groups survive staging
Four MCP feedback reports about the same failure class: the server does no
runtime validation of inputSchema, so a shape mistake was coerced into a
wrong-but-well-formed call, and the error the agent finally saw pointed at
the wrong thing.
- create_voucher / correct_entry: a line naming neither debit_amount nor
credit_amount was `Number(undefined) || 0`-ed into 0/0, and the balance
check reported "debits 0 SEK, credits 0 SEK" for four perfectly balanced
formats ({debit}, {amount, side}, {debitAmount}, signed amount). The line
shape is now checked first, the error names the keys it got and shows a
valid line, and a non-numeric amount is rejected as such (seq 318571).
- match_batch_allocate: kind is the key every guard branches on (direction
vs sign, required id per kind, tenant pre-check on the invoices). With
kind absent none of them fired: an incoming +50 359 SEK payment against
three kundfakturor staged as allocations_kind "supplier_invoice" with zero
invoice checks. A missing or unknown kind is now rejected before any
query, with the id field that goes with each kind (seq 319919).
- categorize_transaction on an already-booked transaction returned the
core's success-shaped object, which fails STAGED_OPERATION_SCHEMA on
strict clients: the agent saw "Structured content does not match the
tool's output schema" and never the reason. It now throws, naming the
existing journal_entry_id (seq 288574).
- reconcile_match: the dry run flattens a pair into one link per outside
row, and the staging rebuild put each back as its own 1:1 pair, so an N:1
group (Skatteverket "Avdragen skatt" + "Arbetsgivaravgift" against one
1630 verifikat, sum exact) reached the executor as N pairs each asked to
settle the whole verifikat: PAIR_NOT_CLOSED on all 18. Links sharing a
verifikat now fold back into one pair, mirroring the existing 1:N fold,
and "No linkable pairs" carries the dry run's skip reasons (seq 292682).
No tools/list payload change: no schema text touched.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013yw62FMXGSzo6icFDiBwP3
* docs(decisions): N:1 reconcile fold at staging
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013yw62FMXGSzo6icFDiBwP3
---------
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
01d903d3a3 |
fix(mcp): link_document_to_voucher marks the inbox item handled; document lists stop misleading agents (#2170)
Three MCP feedback reports (seq 265062, 288474, 288577; three companies) hit the same hole: link_document_to_voucher attaches the document to the verifikat but never stamps the inbox item it came from, and both inbox read surfaces derive "handled" from the inbox row's own link columns, never from document_attachments.journal_entry_id. So an attached document stayed "unprocessed" forever: agents re-saw it as missing underlag (one reporter paged 750 rows to find the ~15 real ones), and one user read the five leftover rows as duplicates and was about to delete the only copies of underlag sitting on posted verifikat. - commitLinkDocumentToVoucher and the bulk twin now stamp invoice_inbox_items.created_journal_entry_id, keyed on document_id, CAS on both link columns, 23505 tolerated (samlingsverifikat). Same shape as the create_voucher + inbox_item_id stamp; best-effort so inbox bookkeeping never rolls back a committed link. - gnubok_list_unmatched_documents returns file_name (same embed list_inbox_items uses) and the extraction's page coverage, so an agent can tell "no total on this document" from "we read 3 of 38 pages" and does not have to fetch each document to learn what it is (seq 265062, 288574). - gnubok_list_transactions_without_documents no longer echoes the column default "uncategorized" on rows that are booked by construction: list_uncategorized_transactions uses the same word for "no journal entry yet", and an agent read the label and tried to re-book an already-booked share-capital deposit (seq 288574). No tools/list payload change: the unmatched-documents item schema is untyped and the category field already allowed null. Claude-Session: https://claude.ai/code/session_013yw62FMXGSzo6icFDiBwP3 Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Pierre Grönberg
|
bac9e01e2d |
fix(arcim): offer the company's own accounts as mapping targets (#2164)
The Fortnox migration's account mapping step builds its target dropdown from BAS_REFERENCE alone: the 1290 standard accounts. Any account a company created outside the standard cannot be selected as a target. Seen on a live Fortnox migration: 3005 "Provisioner inom Sverige" was active in the chart, returned by /api/bookkeeping/accounts, visible everywhere else in the app, and missing from this one list, because BAS defines 3000-3004 and stops there. The plain SIE import at app/(dashboard)/import already used the company's own chart (fetchAccounts(false)), so the two routes into the same AccountMappingStep disagreed about what could be mapped onto. Targets are now the company chart unioned with BAS, deduplicated by account number with the company row winning: its name is whatever the user renamed the account to, and that is the label they look for. BAS stays for standard accounts a first migration is about to create, and a failed chart read degrades to BAS rather than throwing, since an incomplete list still lets the migration proceed while an exception stops it. |
||
|
Mattsson
|
158108ec01 |
fix(bank): keep other companies' accounts out of the EB account picker (#2141)
* fix(bank): keep other companies' accounts out of the EB account picker At one-session banks (SEB) a single BankID consent returns every account the signer can see across all their companies, so a reconnect from company A carries company B's accounts. PR #2116 made those arrive unchecked, labelled and unmirrored; they were still listed in company A's picker and in the connection's account list in settings, which read as "the wrong company's data in my books" (user report, Deepgrid group). - New lib/claimed-accounts.ts: partitionByClaim() splits a connection's accounts on claimed_by_company_id; describeClaimedElsewhere() renders the one-line Swedish summary. Unit-tested, including the legacy double-claim (no flag, stays own) and carried-deselection cases. - AccountPickerDialog: main list, "Markera alla" and the "x av y valda" counter cover own accounts only. Claimed accounts sit behind a collapsed "N konton synkas i <bolag>" disclosure (still tickable: a claim is a strong hint, not proof of ownership). Row markup extracted into renderAccountRow so both lists share it. - BankConnectionStatus: foreign rows dropped from the details list and the "x av y konton synkas" count; one muted summary line instead. No data or callback changes; brand-new never-claimed accounts still list unchecked, since Enable Banking's account resource carries no owner org number. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GY5eTAUFZDoCfbdWoERrsi * fix(bank): close the review and skeptic findings on the claimed-accounts picker Review (CodeRabbit): describeClaimedElsewhere decided "same claimant" on the display name; two companies can share a name. Now keyed on claimed_by_company_id as well, with a test. Skeptics (correctness + regression): - The empty-own-list message asserted "synkas redan i andra bolag" even for a consent with no accounts at all (failed connect, nothing ticked at the bank). Now only when claimed accounts exist; otherwise a plain "inga konton" message. - "Markera alla" stayed enabled but inert with zero own accounts: allSelected is now vacuously true there, so the button disables. - A claimed account ticked inside the disclosure kept counting after the disclosure was collapsed: the disclosure line now names the ticked count so the "x av y valda" counter never exceeds what is visible. - The pending_selection row in settings still counted foreign accounts ("3 konton tillgängliga" beside a picker saying none): now own accounts, with a dedicated line when everything is claimed elsewhere. - Claim flags did not survive an in-place renewal (accountsMetadata is rebuilt without them and the guard skipped seen-on-row accounts), so the sibling's accounts returned to the main list unlabeled on the next reconnect. The callback now re-derives the label from a fresh lookup for accounts that stay disabled here; released claims clear themselves. Two callback tests. Skeptic (compliance) hardening: - partitionByClaim requires enabled === false alongside the flag, so a flagged-but-enabled row (any future writer) can never hide a syncing account. - The sibling company's name is data-ph-masked on the settings summary line and the disclosure line, matching the row label. Declined: CodeRabbit docstring-coverage warning (repo has no docstring requirement; the touched functions carry inline rationale comments). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GY5eTAUFZDoCfbdWoERrsi * fix(bank): keep a re-stamped sibling claim out of the cash-account mirror CodeRabbit round 2: the renewal branch that re-derives the claim label left the account out of guardDisabledUids, so the mirror below still ran upsertFromPsd2 for it. The first connect never mirrored that account (#2116), so a renewal would have planted the sibling's IBAN in this company's cash_accounts and burned a 19xx slot for an account that stays off. Now excluded like a fresh claim; the renewal test asserts only the own account is mirrored. Declined (recorded for the summary): compliance-swarm advisory that the sibling's account metadata reaches the client. Both companies belong to the same signed-in user and the data arrives under that user's own PSD2 consent; the ownership decision is already made server-side in the callback, the picker only renders it. Non-blocking, no cross-user data. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GY5eTAUFZDoCfbdWoERrsi --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
d670fe6663 |
feat(invoices): named payee accounts and per-invoice choice of bank account (#2233)
* fix(enable-banking): read BBAN from AccountIdentification.other and store it on the account Enable Banking has no top-level `bban` key on AccountIdentification: a Swedish BBAN (clearing + account number) arrives as `other.identification` with `other.scheme_name = 'BBAN'`, or in `all_account_ids`. The client typed `bban?: string` and read `.bban`, so the value was always undefined: no connected account ever carried its clearing + account number, and domestic counterparty accounts on transactions were dropped. Type the identifiers per the OpenAPI spec, add extractBban() and pickAccountIdentifier(), read counterparty identifiers through the scheme list (IBAN, then BBAN/BGNR/PGNR, then anything), and store `bban` on StoredAccount from the OAuth callback. The external_id dedup scope stays IBAN-then-uid and is untouched. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * feat(invoices): named payee accounts on cash_accounts with a default per currency A company had exactly one set of payment instructions per invoice currency (company_settings.invoice_payment_accounts), picked by currency alone. A second SEK bank account, or a second bankgiro number, had nowhere to live. cash_accounts is already the per-company bank-account entity. Migration 20260903150000 adds the payee fields (bankgiro, plusgiro, clearing + account number, BBAN, BIC, Swish, foreign routing) plus invoice_payee, a small invoice_payee_defaults table (one default account per currency; one account may be the default for several currencies, a SEK account with an IBAN is the usual EUR payee), and a SECURITY DEFINER mirror that rewrites the legacy map and the SEK bank columns from the default accounts. Every existing reader (PDF, email, reminders, v1, MCP) keeps working; the three writers that only touched legacy columns (PUT /api/settings, v1 settings, MCP update_company_settings) now write through to the default account, so what an agent sets is what the PDF prints. Peppol PaymentMeans is built from the resolver instead of the raw legacy column. bg_pg is dropped (never read or written; NULL on every prod and staging row). Backfill lands only on existing cash accounts (primary, IBAN match, or the only enabled account in the currency). Entries with no target stay in the map as the resolver fallback and get an attach action in settings. New: POST /api/cash-accounts (manual bank account on the next free 19xx), PATCH /api/cash-accounts/[id] payee fields (owner/admin), GET/PUT /api/cash-accounts/payee-defaults. Settings page rewritten as an account list with per-currency defaults. Behandlingshistorik and the full archive cover the new table and columns. Verified on staging: migration applied (11 defaults landed), mirror trigger observed rewriting company_settings from a payee edit. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * feat(invoices): choose which bank account an invoice is paid to, frozen at issue Migration 20260903160000 adds invoices.payment_cash_account_id (FK to cash_accounts, SET NULL) and invoices.payment_details, the payee fields frozen when the account is chosen and refreshed at issue. Resolver: resolveInvoicePaymentAccount / companyWithInvoicePaymentAccount / assertInvoicePaymentAccountForRender take an optional override, and hasRequiredInvoicePaymentAccount reads it from the invoice row, so every surface (PDF, Swish QR, email, reminders, payment confirmation, Peppol, recurring, staged MCP send) prints the frozen payee when one exists and the company default per currency otherwise. Invoices that never chose an account behave exactly as before. Issue paths (mark-sent, send, v1 send, v1 mark-sent, Peppol send, recurring, MCP send and mark-sent) refresh the snapshot from the account as it is at issue; a chosen account that is disabled, un-flagged or unusable for the currency blocks with INVOICE_SEND_PAYMENT_ACCOUNT_INVALID. Writers: dashboard POST/PATCH, v1 create/update and MCP create_invoice accept payment_cash_account_id and validate it against the company's payee accounts (INVOICE_PAYEE_ACCOUNT_INVALID). Credit notes inherit the original's payee; copies carry the choice; preview-pdf renders the chosen account. The editor shows "Betalas till" under the currency when the company has two or more usable payee accounts for that currency. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * feat(invoices): book manual payments on the invoice's chosen bank account Manual mark-paid (dashboard, v1, MCP gnubok_mark_invoice_as_paid) and the booking dialog's proposed lines debited 1930 regardless of which bank account the invoice asked to be paid to. They now resolve the chosen payee account's ledger account (resolveInvoiceSettlementAccount) and fall back to 1930 only when no account was chosen or the row is gone. Bank-transaction matching keeps debiting the account the money landed on and does not filter by the chosen account; between equal-confidence candidates it prefers the invoice that asked to be paid to the landing account. Scores are untouched, so nothing new auto-matches. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * chore(invoices): keep the payload-size and phantom-column ceilings after the payee work Shorten the new gnubok_create_invoice argument description (tools/list payload was 29 bytes over the 60 kB budget), inline the cash-account payee UPDATE/INSERT payloads and the settings select strings as literals so the phantom-column scanner can read their columns, and reuse ACCOUNT_NUMBER_RE instead of a hand-rolled copy. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * fix(invoices): harden the payee model after review (admin-only payee columns, separate payee IBAN, company-scoped FK) Review findings from CodeRabbit, Superagent, the Swedish accounting review and three skeptic passes, resolved in one batch: Schema (both migrations are unshipped and edited in place): - cash_accounts.payee_iban: the printed IBAN is its own column. iban stays the bank identity written by every sync and used to re-pair on reconnect, so a sync can no longer rewrite an invoice instruction or resurrect a cleared IBAN. The backfill copies each currency entry verbatim onto the target account (IBAN match first, then primary), so every invoice keeps printing exactly what it printed before; the bank IBAN is never pushed onto invoices that did not carry one. - Payee columns are owner/admin-only at the database (BEFORE trigger, service role exempt): cash_accounts is member-writable for bank sync, and the SECURITY DEFINER mirror would otherwise have let a member rewrite where customers pay. - Revoking an account as payee or disabling it drops its defaults; deleting a default drops that currency from the map and clears the legacy SEK columns (an admin saying "nothing to print" must not keep printing a closed account). The mirror leaves the legacy SEK columns alone when the map has no SEK entry, so legacy-only companies are never wiped by a mirror run for another currency. - Audit and mirror triggers fire on the same column set; anon and authenticated can no longer execute the trigger-only definer functions. - invoices.payment_cash_account_id is a composite same-company FK with SET NULL scoped to the account column. Code: - Only 19xx bank accounts can be payee: PATCH, the defaults PUT (which now also requires enabled, payee-flagged and usable for the currency), resolveInvoicePayeeChoice, and the mark-paid settlement resolver (which also refuses disabled rows and logs every fallback to 1930). - createManualBankAccount excludes every ledger slot any row already holds (findFreeLedgerAccount treats a manual holder as free; this path inserts). - The legacy settings writers (PUT /api/settings, v1, MCP) write through to the account BEFORE updating company_settings and fail the request on error; the account is written before it is adopted as default so the mirror never sees an empty payee. - snapshotInvoicePayee: dry runs no longer persist; a failed snapshot write blocks issue (INVOICE_PAYEE_SNAPSHOT_FAILED). v1 mark-sent/mark-paid projections carry the payee columns; v1 create validates the payee before the dry-run return and echoes it in the preview. - pickAccountIdentifier: supplementary IBAN wins over a primary BBAN, and non-account schemes (card PANs) are never persisted. - Editor shows the payee select for a single usable account with no default; the booking dialog waits for cash accounts before proposing lines; a failed default write no longer hides a created account. - Behandlingshistorik names the account on created/deleted defaults. - Regenerated skills/accounted-api; MCP argument description trimmed under the tools/list payload ceiling. Declined: clearing legacy columns via a forward migration (the mirror now does it on delete); Swedish review's "show the debit account in the mark-paid UI" (the booking dialog already proposes and lets the user edit the debit line); manual ledger collision (UNIQUE exists, and the create path now rejects it with a clear error); Peppol aligning to the PDF value for companies whose legacy column had drifted from the map (the PDF is the customer-facing document; both now agree). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * fix(invoices): read NEW.invoice_payee only on the cash_accounts branch of the mirror trigger trg_mirror_invoice_payee_defaults fires for both tables; plpgsql resolves record fields per expression, so the combined condition failed with "record new has no field invoice_payee" whenever a default row changed, which took down every pg-real case on the payee tables. The revoke/disable check now sits inside its own TG_TABLE_NAME branch. The MCP settings executor test mocks the payee write-through like the settings route test already does. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * fix(invoices): keep member disables from revoking payee defaults, gate payee on 1920-1999, fit the MCP payload Cycle 3 of /resolve-pr on #2233. Superagent P1: the SECURITY DEFINER mirror trigger deleted an admin's invoice_payee_defaults rows whenever cash_accounts.enabled flipped to false, and enabled is member-writable (the bank picker's "Synkas ej"), so a member could undo an admin's payee decision. The trigger now drops defaults only on the admin-only invoice_payee true -> false revoke; the mirror trigger's WHEN no longer lists enabled. Disabled accounts stay out of the pick lists and the send gate already refuses an invoice that chose one. Applied to staging as the same function + trigger definition and probed inside a rolled-back block: disable keeps the default and the mirrored bankgiro, revoke clears both. pg-real: the admin-guard test ran three expectations inside one withUserContext transaction; the first raise aborted it and the next statement failed with "current transaction is aborted". One transaction per expectation now, and the member case also flips enabled to prove the column stays member-level. Swedish review: payee eligibility was /^19\d\d$/, which admits 1910 Kassa and the 1911-1919 tills. A customer pays to a giro or bank account, so isBankCashAccount, CreateCashAccountSchema.ledger_account and the PATCH route now require BAS 1920-1999; tests cover 1910 and 1919. Unit tests (3/4): the tools/list payload guard read 60 025, then 60 014 tokens after main merged #2166 and #2163 alongside this branch. The ceiling is not bumped and no read on this surface is a demotion candidate, so gnubok_create_invoice drops payment_cash_account_id; agent-created invoices print the per-currency default and v1 REST plus the editor keep the field. Recorded in DECISIONS.md. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * chore(migrations): move invoices_payment_cash_account to 20260903183000 after colliding with main's KPI migration origin/main merged 20260903160000_kpi_monthly_include_reversed_originals while this branch held the same version; identical versions abort the Supabase apply. Staging's schema_migrations row was moved to the new version with the file. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UaZTY21HVN57hJoPXKSLjV * fix(invoices): gate invoice_payee on BAS 1920-1999 at the database, and unblock the typecheck ratchet Cycle 4 of /resolve-pr on #2233, on Emil's go. Swedish review: the 1920-1999 payee rule lived only in the routes. The cash_accounts_payee_admin_only trigger now also refuses invoice_payee on any other ledger (INVOICE_PAYEE_ACCOUNT_INVALID, 23514), whoever writes it, and the backfill only targets giro/bank rows, so a company whose single enabled cash_accounts row is a Stripe clearing account keeps its legacy bankgiro in company_settings instead of landing it on 1686. pg test covers insert and update on 1686 and 1910; the function was applied to staging and probed. Typecheck ratchet: main is red from two merges that landed with failing Checks, and every branch that syncs it inherits the errors. - #2242 added POST(req) calls to the fiscal-periods route test without the route params argument withRouteContext handlers take (25 errors in the file, baseline 23). All 25 calls now pass createMockRouteParams({}). - #2247 made SyncResult.requestedFromDate and historyNarrowed required; the 13 mockedSync results in the enable-banking accounts-route test lacked them. They now carry a fixed date and historyNarrowed: false. Both files' tests pass unchanged in behaviour. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * chore(migrations): move invoices_payment_cash_account to 20260903193000 after colliding with main's party_promotion origin/main merged 20260903183000_party_promotion while this branch held the same version. Staging's schema_migrations row must follow (pending: the Supabase MCP was disconnected at the time of this commit). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
88a5d78594 |
fix(inbox): trace every received mail and file multi-recipient mail once per inbox (#2181) (#2244)
* fix(inbox): trace every received mail and file multi-recipient mail once per inbox (#2181) A mail sent to both the +lev and +ver address of one inbox was read as its first recipient only, and an attachment whose processing threw left no row at all: the webhook answered 200, Resend never retried, and the document was gone with nothing for the user to find. Prod showed both shapes for the reporter (a +lev mail Resend accepted with zero inbox rows, and the second PDF of the +ver mail missing). - The webhook now reads every shared-domain recipient, groups them per inbox, files once per inbox with a company-scoped dedupe key, and resolves contradicting tags (+lev and +ver on one mail) to no hint so extraction classifies. - The per-attachment catch writes an error row instead of only a console line. - One InboundMailReceived behandlingshistorik event per mail and inbox records recipients, tags, hint, conflict and the outcome per attachment (filed, duplicate, rejected, failed). No sender or subject, matching the existing PII rule. - GET /inbound-history?days=30 serves those events, company-scoped, and the inbox workspace shows them under Källor as "Inkomna mejl", each filed row a click away. - The list says how many rows the type filter is hiding, with a click back to all types. - Migration 20260903190000 registers the event type and replaces the (email, attachment) unique index with (company, email, attachment). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CoG2CXf8B33Q5wp8gk4kW4 * fix(inbox): keep addresses and sender-typed tags out of the mail record, and let redelivery heal a transient failure Skeptic pass on #2244, two refutations: - The InboundMailReceived payload carried the recipient addresses and every plus-tag verbatim. An enskild firma's inbox local part is the owner's name, the tag is whatever the sender typed, and processing_history is append-only and outside the erasure path; a numeric tag also tripped the PII validator so the record was silently dropped. The event now carries inbox_id, the documented tags (+lev/+ver), an unknown-tag count and the outcome codes. The history route resolves inbox_id to the company's own address at read time. The DB strip trigger from 20260901110000 covers the new type (and is recreated, since staging skipped that file). - The catch-path error row made a Resend redelivery report "duplicate", so a transient download or storage failure that used to self-heal on retry became permanent. The row is marked transient and a redelivery replaces it; rejections (bad type, too large) stay duplicates. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CoG2CXf8B33Q5wp8gk4kW4 * fix(inbox): cap inbound fan-out, flag a truncated mail history, and name a replaced transient row Review pass on #2244: Superagent (bound the number of inboxes one mail can fan out to: five), CodeRabbit (the history route now returns has_more past 200 rows and the panel says so instead of "every mail"), and the Swedish accounting review (a redelivery that replaces a transient error row names the replaced row on the InboundMailReceived record, so the replacement leaves a trace). The migration comment states why the index swap is not CONCURRENTLY: Supabase branching applies migrations in a transaction. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(inbox): resolve every addressed inbox and record the ones past the fan-out cap CodeRabbit and the Swedish accounting review on #2244: slicing recipient groups before the lookup let five unknown local parts starve a real inbox and left companies past the cap with no trace. Every addressed inbox is now resolved (one cheap lookup each), the first five are processed, and the rest get their own InboundMailReceived record with outcome fan_out_capped, shown in the panel as "not processed". Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * chore(inbox): move the inbound-mail migration past the parties versions merged tonight Main moved party_decision_undo to 20260904000100 and added 20260904000200 (#2257, #2258). A version below prod's head is skipped by Supabase branching, so 20260903190000 becomes 20260904001000 unchanged. Staging re-tracked under the new version. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
b07efcafd4 |
fix(payroll): require jamkning valid_to on every write path (#2058) (#2240)
* fix(payroll): require jamkning valid_to on every write path (#2058) A jamkningsbeslut saved through the v1 API or MCP with a percentage and a start date but no end date was stored and returned 200, yet the engine (isJamkningValid) never applies a beslut without both dates: the payslip and the AGI carried the table tax while the caller believed the beslut was live. One shared validator (lib/salary/jamkning-rules.ts) now requires both dates whenever a percentage is set and checks their ordering. Every write path runs it: CreateEmployeeSchema and UpdateEmployeeSchema, the web POST and PATCH routes, the v1 PATCH route (its private copy is deleted), the MCP create and update executors in employee-commands, and the MCP update tool preflights the merged row at staging time so the agent sees the error before approval. The update paths keep the existing touched gate, so legacy rows stored without valid_to stay editable in unrelated ways. The MCP tool descriptions state that both dates are required for the beslut to apply. scripts/list-incomplete-jamkning.ts lists the existing rows (percentage set, valid_to null) per company, read-only; setting an end date or clearing the beslut is decided per company since either changes the next payslip. Declined: defaulting valid_to to 31 December of the from-year. It matches most beslut but silently changes withholding on rows that today do nothing. Closes #2058 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0161fHpCX3rnWtidwwdGfCdB * fix(payroll): keep the jamkning PR inside the type and tools/list budgets CI on the first push failed on two ratchets this PR itself tripped: - Typecheck ratchet: the three staging tests added here reused the untyped 'agent_chat' actor literal the file already carried, which raised that file's error count above its baseline. They now pass { type: 'user' }. - tools/list payload budget: the first jamkning field descriptions on gnubok_create_employee and gnubok_update_employee pushed the projected catalog to 60 113 tokens against the 60 000 ceiling. The percentage fields keep a one-line "needs both dates or never applied" note; the date fields drop theirs. Also acts on the compliance swarm's GDPR Art.32 note: the read-only lister no longer selects employee names at all (the employee id is what the per-company decision needs), so the script touches no PII. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0161fHpCX3rnWtidwwdGfCdB * docs(mcp): say the jamkning percentage is rejected without both dates CodeRabbit on #2240: "never applied" described the pre-fix engine behaviour; the contract now is that a create or update with a percentage and a missing date is rejected before staging. Same length, so the tools/list payload budget is unchanged. The concurrency finding is tracked in #2256 instead of this PR. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(mcp): keep tools/list under budget after proforma landed on main After merging main (#2254 proforma fields) the projected tools/list measured 60 010 tokens against the 60 000 ceiling with this PR's two jamkning field notes. Per the budget test's own rule, demote a read tool instead of bumping the ceiling: gnubok_list_arsredovisning_versions goes search-only. Versions exist only once a report is rendered for signing or filing, which is the same switched-off iXBRL path as its sibling gnubok_get_arsredovisning_filing_status, already search-only since 2026-09-02. Still reachable via gnubok_call_tool. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
34b677b02c |
chore(ui): retire the Building2 icon app-wide (#2235)
Founder request from the register walkthrough. Suppliers (nav, command palette, empty state) use Truck; company and company-scoped surfaces (active company badge, invite, home signpost, SIE preview, template scopes, TIC workspace and its manifest) use Briefcase; the two bank contexts use Landmark. The extension icon resolver no longer maps Building2; the generated sector definitions follow the manifest. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
b996da60ee |
feat(parties): Förslag från bokföringen, confirmed straight into Leverantörer and Kunder (#2206)
* feat(parties): Kontakter register, suggestion queue, dossier and merge Phase 1's two surfaces on top of the parties substrate: - /parties page: one list with the five-way switch (Alla, Kunder, Leverantörer, Förslag, Bara i bokföringen), search, a 12-month/all period picker, and at most one attention line. Confirmed rows show roles as muted text, rhythm, underlag, dominant account and money. Observed rows are computed and never stored; a generic band keeps unattributed spend visible. - Suggestion queue: a reason per row, hard-key rows pre-ticked, bulk confirm behind one dialog, dismiss on hover, undo on the toast. - Dossier slide-over: Pengar, Bokföring, Vad Accounted vet (facts and identities with source and count), Underlag och verifikat, Historik. - Merge dialog with a visible, swappable survivor and undo. - API: GET /api/parties, GET /api/parties/[id], POST suggest, decide, decide/undo, merge, merge/undo (withRouteContext, Zod, 15 tests). - Migration 20260903090000: decide_parties snapshots the reason it clears; undo_party_decisions reverses confirm/dismiss within 30 days; decision kind 'undo'. - The pipeline runs after SIE import and provider migration (non-blocking) so a migrant's register is full on arrival. - Nav entry under Register; sv/en strings. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(parties): pass explicit interpolation values to next-intl next build's type check rejects a typed interface where the translator wants an index-signature record. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(parties): retry label on the load-failed state Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(parties): hard keys for companies without org number, readable names, look-alikes at read time - get_ledger_key_evidence dropped every document for a company whose own org number is NULL (the self check compared against NULL). Replaced in 20260903100000 with a coalesced comparison; pg test covers it. - Display names come from the printed name on documents, otherwise from the voucher text with the AP/AR prefix and supplier number removed. - Look-alike parties (same core, or one core extending the other by whole words: Fortnox / Fortnox Finans) are detected when the register is read, never stored, and feed the Dubblett? chip and the merge dialog. - Queue shows Intäkt beside Kostnad; dossier hides zero money rows and formats bankgiro/plusgiro; merge dialog cancels with Avbryt; no synchronous setState inside effects. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * feat(parties): link every new supplier and customer to a party on write The backfill covered the rows that existed on 2026-09-02; 108 rows created since had no party and never reached the register. A BEFORE INSERT/UPDATE trigger on customers and suppliers now calls ensure_party on every write path at once: find-or-create by org number inside the company, never by name; a private customer gets a kind=person party without any number; a nameless row stays unlinked; a foreign party id is refused with the same error as the composite foreign key; a link to a merged party follows the chain to the survivor; the clear that ON DELETE SET NULL performs is kept. ensure_party lets the trigger act for the row's owner (pg_trigger_depth() > 0); the RPC path is unchanged. The migration also links the rows created since the backfill. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(parties): dossier hides dismissed parties and follows merges to the survivor The register hid archived parties while the dossier still served them by id, and a merged party's dossier pointed at a dead row. Superagent P2 on #2206; three unit tests. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * chore(parties): move the role-link migration past main's 20260903110000 Two files with one version would collide in schema_migrations. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * feat(parties): confirm suggestions into Leverantörer and Kunder, no third noun Founder decision after the walkthrough: users know two words. The page becomes the queue 'Förslag från bokföringen' with 'Bara i bokföringen' beside it; the Kontakter nav entry and the Alla/Kunder/Leverantörer views go. Each suggestion shows what it becomes (Blir), read from the ledger side and changeable per row; confirming calls promote_parties, which creates the supplier and/or customer row from the party's facts, never a duplicate, and is undoable for 30 days through undo_party_promotions (the created rows are archived, the party returns to the queue). Leverantörer and Kunder carry the one attention line that leads here. The dossier offers Lägg upp som leverantör / som kund. Migration 20260903130000, 5 pg tests, route and unit tests updated. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(parties): write bankgiro and plusgiro the way the supplier form does Identities are stored as digits; suppliers carry 5317-0900. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * chore(parties): move the four queue migrations past main's 20260903170000 Main merged 20260903120000_skattekonto_transactions_realtime_publication with the same version as the role-link trigger; the preview database refused the duplicate key. All four now sit after main's newest so the set applies in one ordered run on prod. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
a48508e5b0 |
feat(dimensions): show the value's name after picking, and let an unused custom dimension be deleted (#2219) (#2255)
* feat(dimensions): show the value's name after picking, and let an unused custom dimension be deleted (#2219) Two things from the same Discord report, both in bookkeeping from the transaction view: 1. After picking a kostnadsställe the field showed only the code ("1"). DimensionCombobox now writes the value's full name under the field once a code is committed, exactly as AccountCombobox does for the account name (looked up in the full registry so an archived code stays readable). The input text itself stays the code: the blur/revert logic keys on it. 2. A self-created dimension could not be removed at all: the DB already allowed it (enforce_dimension_registry_guards lets a non-system dimension go when no posted/reversed line carries its number, and the value retention trigger fires on the cascade), but no route or UI asked. New DELETE /api/dimensions/[id]: 400 DIMENSION_SYSTEM_DELETE for kostnadsställe/projekt, the guard's own Swedish P0001 verbatim as 409 DIMENSION_REFERENCED, 404, and a happy path; the register gets a quiet "Ta bort dimension" link for the active custom dimension behind a DestructiveConfirmDialog. Keys added to sv and en. Closes #2219 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VnConrmMCxJRQ5kfiPPWyy * test: satisfy the TypeScript ratchet for two test files main inherited from #2247 and #2242 accounts-route.test.ts built SyncResult literals without the requestedFromDate / historyNarrowed fields #2247 added (vitest does not typecheck, so it passed locally); fiscal-periods route.test.ts got two more one-argument POST(req) calls from #2242 in a file already at its ratchet baseline. Both files now typecheck; the ratchet runs clean. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VnConrmMCxJRQ5kfiPPWyy --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
5eac2a492c |
fix(enable-banking): stop reading every ASPSP_ERROR as a too-wide window (#2202) (#2247)
* fix(enable-banking): stop reading every ASPSP_ERROR as a too-wide window (#2202) ASPSP_ERROR is Enable Banking's generic wrapper for any upstream bank failure, so "history window beyond the PSD2 limit" and "the bank is refusing right now" arrived as the same string, and every rejection walked the whole 90/60/30 narrowing ladder: one user click cost up to five upstream calls against a bank that was already saying no, the failure surfaced with "förnya anslutningen" advice that fixes nothing, and a sync that did narrow was reported as complete. What the account has accepted before is the signal that tells the two apart. sync.ts now records the widest window (days before date_to) each account's bank has answered, on accounts_data as accepted_history_days (no migration; persisted by the same write-back as dedup_scope). On a rejected window: no wider than that = the bank is unavailable, stop after one call; wider = one retry straight at the accepted width, then stop. Without a record (first sync, legacy rows) the ladder runs as before, but its exhaustion is now AspspUnavailableError too. The web sync route maps that to 503 BANK_UNAVAILABLE with copy that says the connection does not need renewing and leaves the row alone; the agent path keeps the contract code BANK_SYNC_FAILED but no longer persists renewal advice. getAllTransactionsWithRaw returns the requested and the effective date_from plus a narrowed flag; the sync result and the /sync response carry them (history_from), and the settings toast says from which date the history is complete when the bank cut the window. Not done: a per-account backoff for the user-triggered route (the agent path already has the 15-minute lease from #2165), and using the envelope's `detail` field (one sample, identical to a width rejection). Closes #2202 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VnConrmMCxJRQ5kfiPPWyy * docs(decisions): carry the batch's decision lines (#2237, #2203, #2214) here --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
3918ff6620 |
fix(customers): make country ISO-2 everywhere and check it against the customer type (#2241)
* fix(customers): make country ISO-2 everywhere and check it against the customer type (#2025, #2028) customers.country and suppliers.country were read as ISO codes by the periodisk sammanstallning (SKV 5740), Peppol and the provider importers but written as English names by the customer form and the v1 API, so a correct German customer produced GERMANY811234567 in the SKV file plus two false warnings, and an EU customer saved with land Sverige got reverse charge with nothing objecting until after the invoice was sent. - lib/vat/country-codes.ts: one helper that normalises codes and the Swedish/English names the writers used to store, the country-vs-type rule (swedish_business = SE, eu_business = EU member other than SE that matches the VAT prefix, non_eu_business = outside the EU), and the reverse-charge country gate. - Writers: customer form and supplier form get a country select; internal REST, v1 REST, bulk-create, MCP create/update, CSV/Excel import and the provider migration mapper normalise to a code and refuse unknown text; the consistency rule is a form error and an API 400 (CUSTOMER_COUNTRY_MISMATCH on update). An omitted country is SE for Swedish types, derived from the VAT prefix for eu_business, required for non_eu_business. - vat-rules.ts: getVatRules and friends take the country as a third argument and grant reverse charge only for an EU country other than SE; every invoice/sales-order/MCP call site passes customer.country. - periodisk sammanstallning reads legacy names through the same helper. - Migration 20260903170000: normalize_country_code() SQL twin, country_raw rollback column on both tables, backfill of every non-code row; unknown text is left as-is. pg-real test for the function. Closes #2025, closes #2028 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01D5EmmndLyDCmY5NHYAvYkE * fix(customers): keep reverse charge for defaulted-SE EU rows, gate the country rule on the fields it reads, fix build Skeptic and CI findings on #2241, one pass: - Migration step 4: eu_business rows whose country was null or only the old writer default (SE) while the VAT number names another EU member take the country from the prefix. The pre-2026-09 rules granted reverse charge on type + VIES validation alone, so these rows invoiced at 0% and would have flipped to 25% on the next invoice. country_raw = '' marks a null origin; rollback uses nullif(country_raw, ''). - countryPermitsReverseCharge refuses SE only: a VIES-validated number outweighs a non-EU address (Swiss company registered in DE, Monaco with a FR number, Northern Ireland XI). - checkCountryConsistency: an eu_business outside the EU VAT area is accepted when the VAT prefix is an EU-trade registration (incl. XI); Monaco maps to the FR prefix. - Internal PATCH, MCP update and the commit executor judge the country rule only when customer_type, country or vat_number is part of the update, so a contradictory legacy row can still change its email (v1 already did). - Webshop-order customers get the order's billing country; spreadsheet import derives a missing country from the type and flags contradictions (parser row error + execute schema refine). - Build: v1 [id] route typed the existing row through a narrowed alias (never) and passed messageSv/messageEn the v1 error context lacks; the self-billed customer projection lacked country. - Checks: regenerated skills/accounted-api (customer example country SE). - New parity test holds the migration's SQL name table to the TS table. - DECISIONS.md: correct migration version and the revised rule. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01D5EmmndLyDCmY5NHYAvYkE --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
80b87c55fc |
fix(skatteverket): treat AGI kvittens gateway refusals as errors and name the connector operator (#2234)
* fix(skatteverket): treat AGI kvittens gateway refusals as errors and name the connector operator The AGI kvittens cron bucketed ACCESS_DENIED as apigw_config with a warn-once suppression because the APIGW client was known to lack the AGI hantera subscription in Utvecklarportalen (#963). With that subscription being put in place (#2226), a gateway refusal is a regression and belongs in the ordinary error path, so the bucket, its "known configuration gap" comment and the apigwConfig response field are gone. The connector-mode gateway-refusal message said "kontakta supporten"; hosted is now itself a Connect installation for the canary companies, so the message names the connector operator by host instead. Refs #2226 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0159Mi1sTAHfUDZkSysntYz2 * fix(skatteverket): keep the ACCESS_DENIED code in the kvittens cron body Skeptic finding on PR #2234: the generic error path ran the gateway refusal through getErrorMessage, whose Swedish keyword heuristic misses the gateway wording and collapsed it into "Något gick fel". Echo the machine-readable code instead, as the expired_token and grant_revoked rows already do; the full guidance stays in the error log. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0159Mi1sTAHfUDZkSysntYz2 --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
cb39cded81 |
fix(payroll): declare AGI for the payout month, not the run's period month (#2191) (#2228)
Arbetsgivardeklarationen is filed for the calendar month the pay went out (kontantprincipen), so a run for August paid on 25 September belongs to redovisningsperiod 202609. The generator, the submit route, the run page and the run header all took run.period_year/period_month instead, and three PATCH paths refused any payment date outside that month, which made lön i efterskott impossible to set up at all. - lib/salary/agi/reporting-period.ts: one dependency-free helper (agiReportingPeriod) derives the period from payment_date, falling back to the run period only when the date is missing. - generate-declaration.ts: XML Redovisningsperiod, the agi_declarations lookup/insert and the sanity warnings key on the payout month. New AGI_PERIOD_CONFLICT (409) refuses to overwrite another live run's declaration for the same payout month; corrections still replace. - submit route, run page (AGI panel, submission hook, tax-payment fetch, XML filename) and RunHeader use the helper; the header says "AGI redovisas för 2026-09 (utbetalningsmånaden)" whenever the two differ. - The in-period payment-date guard is lifted in the dashboard PATCH, lib/salary/update-run.ts (MCP staged tool + pending-ops executor) and the v1 PATCH, plus the RunHeader min/max; its only stated reason was the period-keyed AGI. Generated API skill reference updated. Existing agi_declarations rows keep their stored period: a declaration already filed under the earned month is a correction with Skatteverket, not a re-key. Rule verified against Skatteverket's guidance on redovisningsperiod (kontantprincipen). Closes #2191 Claude-Session: https://claude.ai/code/session_01QPQLwHNEiQfiCNLSMzXMiQ Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
d900fea1a8 |
feat(connect): Skatteverket data calls through the connector with a per-company canary (#2209)
CONNECT_SKV_CANARY_COMPANIES mirrors the bank canary: the listed companies' user-token Skatteverket calls (skattekonto, moms, AGI) go through the connector's data proxy while this installation still has its own credentials and keeps refreshing the tokens on its own OAuth client. Callers without a company id (OAuth start, token exchange, environment reporting) keep the plain rule: own credentials win. This is how hosted moves its Skatteverket traffic to Connect a few companies at a time before dropping its own keys. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> |
||
|
Mattsson
|
fefef038c5 |
fix(sales-orders): pin the sales_order_items embed FK and teach the embed guard composite keys (#2207)
* fix(sales-orders): pin the sales_order_items embed FK and teach the embed guard composite keys Migration 20260902180000_sales_orders_hardening added a composite (sales_order_id, company_id) foreign key from sales_order_items to sales_orders next to the original single-column one. PostgREST then saw two relationships and answered every `items:sales_order_items(*)` embed with HTTP 300 / PGRST201, so kundorder list, detail, create and the MCP list tool all failed on prod and staging with "Oväntat serverfel". - Hint the three embeds with `!sales_order_items_sales_order_id_fkey` (route, load service, MCP list tool). - scripts/checks/ambiguous-embed.mjs only parsed single-column `FOREIGN KEY (col)`, which is why the ratchet reported 0 for this pair. It now reads composite column lists (named or default constraint name) in both CREATE TABLE and ALTER TABLE, derives the same 17 ambiguous pairs prod's pg_constraint reports, and flags all three shipped sites on main. - Unit tests for the composite shapes: alongside a single-column key, replacing one, and inline in CREATE TABLE. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Q7xJQL2aZo6iRHxCZNntUq * fix(checks): drop composite embed edges when DROP COLUMN removes a member column Postgres drops every foreign key a column takes part in, so the ambiguous-embed parser must release a composite edge (and its constraint name) when one of its columns is dropped, not only the single-column key. Otherwise a later migration would keep a pair armed for a relationship that no longer exists and reject valid embeds. Regression case added. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Q7xJQL2aZo6iRHxCZNntUq --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
4c6feea64d |
feat(connect): bank sync through the connector operation, with a per-company canary (#2205)
* feat(connect): bank sync through the connector operation, with a per-company canary In connector mode the enable-banking sync no longer pages Enable Banking on the instance: it calls POST /api/connect/bank/sync on the hosted service with the session id it holds and the account, and receives booked, normalized rows plus the raw provider pages to archive. Everything downstream is shared with the direct path (stored external ids computed here from booking_date, amount and the account scope; ingest; archive; balance refresh), so a company that moves to the connector produces byte-identical keys. A 410 from the service maps onto the same SessionExpiredError the direct path throws. bankConnectorMode(companyId) gains CONNECT_BANK_CANARY_COMPANIES: listed companies use the connector even while the installation has its own Enable Banking credentials, which is how hosted Accounted moves its bank sync to Connect a few companies at a time before dropping its keys. The contract package gains the bank sync request/response schemas (2026-09-03). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Signed-off-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> * fix(connect): calendar-valid dates, body read inside the timeout, service origin as the default Review follow-ups on #2205. The contract validates date_from, date_to and booking_date with z.iso.date() (2026-02-30 and an empty booking date are refused; the installation derives its stored keys from booking_date). The connector sync reads the response body inside the timeout window so a service that stalls the body cannot hold the sync open. DEFAULT_CONNECT_BASE_URL now names the connector service (connect.accounted.se), which is where the sync operation exists; the hosted app's copy of the connector routes is legacy and hosted Accounted itself sets GNUBOK_CONNECT_URL explicitly. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Signed-off-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> --------- Signed-off-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
1c82baf553 |
feat(invoices): offert (quote) document type with own OF-series, decisions, conversion, MCP and v1 (#2163)
* fix(invoices): reminders, AR ledger, AR reconciliation and deadlines only read fakturor The overdue-reminder run, the kundreskontra, the 1510 reconciliation and the deadlines page selected invoices by status alone. A sent proforma past its due date was chased with a betalningspaminnelse and flipped to 'overdue', and it appeared as a receivable. All four now filter document_type = 'invoice', which is also the precondition for adding quotes (offert): a quote carries a date but never a receivable. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * feat(invoices): offert (quote) document type with its own OF-series, decisions and conversion Adds document_type 'quote' with valid_until, quote_status (open / accepted / declined; expired is derived from valid_until, never stored) and quote_decided_at. Quotes are numbered OF-nnn at insert from company_settings.next_quote_number via generate_quote_number(), the same pattern as delivery notes, so a declined quote never leaves a hole in the F-series the way a proforma does. The column next_quote_number already existed on prod and staging without a migration; the migration adopts it. Engine: build-invoice-write writes the quote columns and keeps remaining_amount at 0; the draft editor refuses accepted or declined quotes; PATCH refuses changing a quote's or delivery note's document type since the number belongs to the series; mark-paid refuses quotes. New POST /api/invoices/[id]/quote-status records the decision and locks once an invoice exists. Conversion is extracted into lib/invoices/convert-to-invoice.ts (one implementation for the route and the MCP staged commit, which had drifted): a converted quote stays and flips to accepted, the invoice links back via converted_from_id and gets its due date from the customer's payment terms; a declined or already invoiced quote is refused. next-number previews the OF-series for quotes. Migration applied to the staging branch and registered as 20260902140000; the pg test runs in CI (pg-real). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * feat(invoices): quote PDF, email and filename surfaces The customer-facing surfaces get a quote sibling for every proforma branch: PDF title OFFERT / QUOTE with Offertdatum and Giltig till instead of the due date, a notice that the document is not an invoice or a payment request, and no payment box, OCR, bankgiro, Swish, QR or payment link. The email says the quote is attached and valid until the expiry, drops the payment details and pay-online button, and asks about the quote rather than the invoice. Filenames read "Offert nr OF-001". Seller VAT number and payment accounts are skipped for quotes as for proformas: a quote is not a faktura under ML 17 kap. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * feat(invoices): offert in the editor, list and detail pages Editor: "Offert" document type with a required "Giltig till" field (default today + 30 days) in place of the due date; the wire body mirrors it into due_date so the shared schema is satisfied. Payment link, ROT/RUT, periodisering and the bank box are already gated on real invoices. The type cannot be switched on an existing quote (its OF-number belongs to the series). List: an Offerter tab beside Proforma, "Ny offert" in the split button, and a status column that shows the decision or the derived expiry: Utgången and Avböjd are exception chips, Öppen and Accepterad muted text. Detail: Acceptera and Skapa faktura in the header, Avböj in the overflow menu; an expired quote asks before accepting or invoicing (bypassable); once an invoice exists the page links to it as Fakturerad and hides the decision actions. Strings in both sv and en. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * feat(mcp,v1): expose offert on the MCP tools and the v1 REST surface MCP: create_invoice takes document_type quote with a required valid_until and allocates the OF-number at insert; the convert tool keeps its id and accepts quotes with the registry refusal codes; new set_quote_status; list_invoices and get_invoice expose valid_until and the effective quote status, including a derived expired filter. The tools/list payload stays under its ceiling without a ledger change. The MCP staged convert now uses the shared converter. v1: POST /invoices/{id}/quote-status (registered in the endpoint registry, scope map and route loader), valid_until and quote_status in the list, create and detail shapes, and a quote_status list filter. Skill atoms mention offert. Decision log lines for the own number series, derived expiry, accepted-not-cancelled conversion and the header action layout. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * test(invoices): pass route params and period id in the new quote tests Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * refactor(invoices): literal update payloads in the converter so the phantom-column guard can read them Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01W45yD8NfQ97JhyYaXpzN56 * fix(invoices): close the quote review findings in one pass Skeptics (correctness, compliance, regression) and CodeRabbit on #2163: - quote_status is no longer a write-builder output, so a v1 PATCH or MCP update_invoice can never reset a recorded accept/decline; new quotes are opened by the invoices_quote_defaults trigger (20260902141000), which also keeps due_date and valid_until equal. v1 PATCH and the MCP update executor now use the shared editable-draft predicate. - One live invoice per converted source, enforced by a partial unique index; the converter maps 23505 to INVOICE_QUOTE_ALREADY_INVOICED and both quote-status routes compare-and-set on the decision they read. - MCP-created quotes carry remaining_amount 0; mark-paid, transaction match and voucher link refuse non-invoices on the MCP staging tools, the executors and the dashboard link route. - Conversion of a foreign-currency source refetches the rate for the conversion day (ML 8 kap 21-23 paragraphs) and fails closed without one; 0-day payment terms mean due on receipt. - bulk-create refuses quotes per item; list_invoices rejects a quote_status filter combined with another document_type; an omitted document_type on PATCH means unchanged. - attention, push notifications, open-AR count, FX revaluation, year-end and accrual auto-detect and bank-match suggestions only read fakturor. - Quote PDF and email print Summa / Total instead of Att betala. - Regenerated skills/accounted-api for the new v1 endpoint. Declined with reasons in DECISIONS.md: NOT VALID + VALIDATE and CONCURRENTLY on the migrations (repo precedent, 13.8k rows, transactional apply); re-validating VAT treatment at conversion (the converted invoice is a draft the user reviews; follow-up). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0111fYAUxKtpxU1BHiBioqzs * fix(invoices): second review round: migration versions, order links, batch allocation, races - Migrations renamed to 20260902220000 / 20260902221000: #2166 shipped its own 20260902141000 to prod while this PR was in review and prod's head moved past both files; below-head versions are skipped by branching, which would have left the quote trigger off prod. Staging rows renamed. - Quote lines never carry sales_order_item_id (an offer must not count as invoiced kundorder quantity); the converter carries a proforma line's order link onto the invoice. - Converter compare-and-sets the source (proforma cancel, quote accept): a concurrent cancel, proforma-to-order conversion or decision removes the orphan invoice with INVOICE_CONVERT_SOURCE_CHANGED instead of a second document for the same sale. - MCP set_quote_status gets the same compare-and-set as the HTTP routes; 0-row updates report INVOICE_QUOTE_CHANGED_CONCURRENTLY everywhere. quote-status (dashboard, v1, MCP) accepts valid_until so an expired sent quote can be reopened, as the docs promised. - MCP mark-paid refuses only quotes, parity with the dashboard route (a sent proforma marked paid is a supported prepayment record). - Batch allocation (dashboard route and MCP tool) refuses non-invoices before the RPC, which gates on status alone. - Customer AR drill-down, v1 customer open invoices and archive guard, and the calendar feed read fakturor only. - Draft quote PDF says "UTKAST" instead of "not a valid invoice"; the editor locks the document type on existing quotes and delivery notes. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0111fYAUxKtpxU1BHiBioqzs * chore(invoices): use roundOre in the quote MCP summaries and FX test after main tightened the guard baseline Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0111fYAUxKtpxU1BHiBioqzs * fix(invoices): third review round: atomic decision lock, viewer gate, lookup errors, quote payment terms - 20260902222000: BEFORE UPDATE trigger locks an accepted quote while a live converted invoice exists (the compare-and-set in the three decision writers could still be beaten by a conversion landing in between); the routes and the MCP tool map the raise to 409 INVOICE_QUOTE_ALREADY_INVOICED. generate_quote_number now also requires a non-viewer membership so a viewer's session token cannot burn OF-numbers through PostgREST. - Converter checks quote eligibility before the Riksbanken call and treats a failed company_settings read as a failure instead of a 30-day default. - Re-sending the same decision keeps quote_decided_at (idempotent). - gnubok_find_voucher_candidates_for_invoice refuses non-invoices like its write sibling; the dashboard link route surfaces a failed lookup. - Late-fee and credit-term texts never print on a quote. Applied and registered on staging; pg tests added. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0111fYAUxKtpxU1BHiBioqzs * fix(invoices): review nits: fail-closed batch lookup, dry-run expiry, quote heading, quote-date CHECK - match-batch surfaces a failed document lookup instead of allocating. - v1 quote-status dry-run preview carries the new valid_until. - Quote PDF heading reads Offertinformation / Quote information. - 20260902222000 also pins the date invariants the trigger maintains as a CHECK: a quote always has valid_until = due_date, nothing else has one. Applied on staging. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0111fYAUxKtpxU1BHiBioqzs --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
8c8996773f |
chore(connect): provider-host ratchet in check:guards, drop the dead Arcim gateway client (#2178)
* chore(connect): provider-host ratchet in check:guards, drop the dead Arcim gateway client Two boundary chores from the Connect plan. (1) A per-file ratchet in scripts/checks/no-new-antipatterns.mjs over files under lib/, app/ and extensions/ that name a provider API host (Enable Banking, Skatteverket, Qvalia, Fortnox, Visma, Briox, Bjorn Lunden, Bokio, Bolagsverket, TIC, Meta, Gmail). The 22 files that do so today are grandfathered in the baseline; a new one fails the guard with the connector routing as the remedy, and the set may only shrink as upstreams move behind the connector. (2) The client for the retired Arcim Sync gateway (extensions/general/arcim-migration/lib/ arcim-client.ts) is deleted with its test: provider-client.ts replaced it and nothing else imported it. The --update rewrite also locks in the lower naive-ore-round count (620 to 617) that main already reached. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MMUUom4fUk6zi4xYZSSfat Signed-off-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> * chore(guards): provider-host ratchet is case-insensitive and skips colocated .test.tsx; document the own-credentials exception Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Signed-off-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> --------- Signed-off-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
c0818bb2d2 |
feat(sales-orders): kundorder with partial delivery and partial invoicing (#2166)
* feat(sales-orders): kundorder with partial delivery and partial invoicing Adds sales orders (kundorder) as their own non-ledger document between agreement and invoice, for companies that deliver or invoice in parts. Schema (20260902130000): sales_orders + sales_order_items with RLS via user_company_ids(), OR-<n> numbering RPC (membership-gated, no anon execute), company_settings.sales_orders_enabled UI gate, and back-links invoices.sales_order_id / invoice_items.sales_order_item_id. The invoiced quantity per order line is DERIVED from the linked invoice lines on non-cancelled, non-credited invoices and enforced by a BEFORE trigger, so no counter can drift and a credited invoice frees its quantity. Header status is draft / confirmed / completed / cancelled; completion is kept by DB triggers from the same derived quantity. Delivery and invoicing progress are derived per line, never stored as status. Service + API: lib/sales-orders (create/update with id-preserving line replace, transitions with compare-and-set, cumulative delivery registration, invoice-from-order through buildInvoiceWriteData so booking stays in the engine, proforma -> order conversion), routes under /api/sales-orders and /api/invoices/[id]/convert-to-order, structured SALES_ORDER_* error codes, archive classification of the new tables. The invoice editor round-trips sales_order_item_id so a draft edit cannot drop the link; GET /api/invoices gains ?sales_order_id=. UI: /sales-orders list, create/edit form reusing the invoice line conventions, detail with deliver and create-invoice dialogs and linked invoices; nav row behind the settings toggle; the webshop row is relabelled webshop_orders; "Skapa order" on proformas. MCP (20260902141000/141001): list/get reads plus four staged writes (create, transition, register delivery, create invoice from order) whose executors call the lib services; op types added to the pending operations CHECK. Tests: route tests for every route (401/400/404/happy), service unit tests, executor and tool tests, and tests/pg/sales-orders.pg.test.ts (16 cases, green on staging) covering RLS, numbering guards, the over-invoice trigger incl. release on cancel/credit and cross-company refusal, the quantity floor, and completion maintenance. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RQW7mXvbAPgjUHq7dSEamr * fix(sales-orders): harden kundorder after skeptic and security review Resolves every finding from the PR #2166 review pass in one batch. Order link integrity: replaceInvoiceItems now refuses a line set that drops an existing sales_order_item_id (INVOICE_UPDATE_DROPS_ORDER_LINK), closing the MCP update_invoice header-only edit and the v1 PATCH path that severed the link and freed the quantity for double invoicing. The update_invoice re-fetch, gnubok_get_invoice and the v1 item projection now carry sales_order_item_id so well-behaved clients round-trip it. Quantity math: derived remaining/invoiced quantities are rounded to six decimals and compared with an epsilon (roundQty, qtyGreater) so a float remainder such as 0.5999999999999996 can neither refuse the final partial invoice nor land as an invoice quantity; duplicate explicit picks are summed before validation. Leveransdatum: per-line last_delivery_date (migration 20260902160000); an invoice takes the latest date over the lines it covers and only when the covered quantity was delivered, never the header date and never for an advance invoice (ML 17 kap 24 p.7, FX anchor per ML 8 kap 21-23). VAT drift: the order stores the customer type and VAT-validation flag its lines were priced under; invoicing refuses with SALES_ORDER_CUSTOMER_VAT_CHANGED when they differ, and re-saving the order re-validates the lines. Customer and currency are frozen once invoices exist. Tenant and role gates: composite FK (sales_order_id, company_id) ties a line to its parent's company (Superagent P2); aa_enforce_company_writer_role on both tables so a viewer cannot write through the browser client. Proforma -> order refuses proformas with ROT/RUT, periodisering or negative-quantity lines instead of dropping those fields. RESTRICT FK errors on delete map to SALES_ORDER_LINE_LOCKED / SALES_ORDER_HAS_INVOICES. Also: schema-guard literal payloads in lib/sales-orders (ceiling +2 with reason), regenerated skills/accounted-api (sales_order_item_id on invoice items), pg tests for the composite FK, the viewer gate and the new columns, unit tests for every changed path. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzFmmH82hCJNmZbPqycDiW * fix(sales-orders): resolve CodeRabbit round on PR #2166 Quick wins from the review, all in one pass: - replaceInvoiceItems fails closed when the invoice_items snapshot cannot be read (it is both the restore source and the input to the kundorder link guard); the guard branch is explicit in both PATCH routes. - Cumulative delivery registration carries an optimistic predicate on the quantity it read, so two concurrent registrations cannot regress each other; DELETE of an order keeps its allowed status in the predicate and answers a conflict when zero rows match. - Business dates (order date, delivery date, invoice date) default to the Europe/Stockholm calendar day (todayIsoStockholm), never UTC: the delivery date is also the Riksbanken rate anchor. - The invoice-from-order executor treats an event emit failure as non-blocking: the draft already exists. - sales_order_items are archived through their parent with the order currency denormalised, like invoice_items. - Proforma "Skapa order" tolerates a 2xx without a parsable body; the settings toggle refreshes the server-rendered nav. - List route doc states that q matches the order number (customer names are matched client-side). Declined (out of scope for this PR): moving header + line writes and the delivery loop into transactional RPCs (same PostgREST pattern as the invoice PATCH path, tracked as a follow-up), the MCP approval handler's error message shape (pre-existing code outside this change), and the docstring-coverage warning (no repo convention). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzFmmH82hCJNmZbPqycDiW * fix(sales-orders): move hardening migration off a colliding version; archive contract; ceiling - 20260902160000_sales_orders_hardening.sql collided with main's 20260902160000_parties_substrate.sql after the third sync; renamed to 20260902180000 and made idempotent (DROP ... IF EXISTS before each ADD CONSTRAINT) so a preview branch that applied it under the old version replays it cleanly. Staging's schema_migrations row renamed. - sales_order_items goes back to a direct archive dump: the coverage contract (tests/pg/full-archive-coverage.pg.test.ts) requires it for a table with its own company_id; the currency lives on the parent order one file over, joined by sales_order_id. - Scanner ceiling re-baselined after merging main (parties phase 1): 397. - v1 PATCH test queues a real empty invoice_items snapshot now that replaceInvoiceItems fails closed on an unreadable one. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzFmmH82hCJNmZbPqycDiW * fix(sales-orders): drop the composite FK before its unique index on replay The idempotent guard in 20260902180000_sales_orders_hardening.sql dropped the unique (id, company_id) before the FK that depends on its index, so the preview branch replay (which had applied the file under its former version) failed with SQLSTATE 2BP01. Order swapped; replay verified on staging. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XzFmmH82hCJNmZbPqycDiW --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
b68c082ef5 |
feat(bank-sync): close the F2 report: gap backfill, consent and paused chip states, agent-triggered sync (#2165)
* fix(bank-sync): cron backfills the gap since the last successful sync The daily incremental sync always asked the bank for the last 7 days. Any pause longer than that (a lapsed subscription paid again, a consent renewed after expiry, an outage) silently lost the days in between: the connection came back, looked healthy, and the missing transactions never arrived. The lookback now widens to cover the gap since last_synced_at plus one day of overlap, capped at the 90-day PSD2 limit, and a gap of a month or more asks for strategy=longest like the manual sync route does. Dedup via external_id makes the overlap harmless. First syncs keep their 90-day path. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QdcDV7CngLkWUvfKSsxFhS * feat(bank-sync): chip warns seven days before a bank consent expires The transactions-page chip only reacted once a connection was already dead (expired/error) or had gone stale. A consent that is about to end looked healthy until the morning it stopped syncing. New "expiring" state when a live connection's consent_expires is within seven days, the same threshold as the consent-expiry email in the sync cron. Precedence: attention, expiring, stale, healthy. getChipState moves to lib/transactions/bank-sync-chip-state.ts so the precedence is unit-tested; the component keeps the rendering only. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QdcDV7CngLkWUvfKSsxFhS * feat(bank-sync): chip says paused when the subscription lapsed The daily cron filters connections by the bank_sync capability, so a company whose trial or subscription ended keeps status=active rows with a frozen last_synced_at. The chip read that as "stale, check the connection", which sends the user to re-authorise a connection that is perfectly alive. 56 of 191 active connections on prod were in this state on 2026-09-01. New "paused" state, ranked above everything else, when the company lacks bank_sync: hosted points at billing, self-host at the connector key, the same split BankSyncNowButton already makes. getChipState takes an options object so the clock stays out of render (react-hooks/purity). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QdcDV7CngLkWUvfKSsxFhS * feat(api): agent-triggerable bank sync in v1 and MCP Closes the first wish in the F2 report: an integration could read bank data but never refresh it. New POST /api/v1/companies/{id}/bank-connections/ {connectionId}/sync and MCP gnubok_sync_bank, both on a shared runner (extensions/general/enable-banking/lib/trigger-sync.ts). Cost is bounded structurally, not by policy: the window is never caller-controlled (the cron's gap-aware 7 to 90 day lookback), a connection synced within 15 minutes answers BANK_SYNC_COOLDOWN with next_allowed_at (429 + Retry-After on v1; synced=false in-band on MCP so the agent reads on instead of retrying), and a failing connection is throttled per process by attempt time. A dead session is flipped to expired with a remediation that hands the user the connect link: no API call revives a consent. Gated on bank_sync like gnubok_connect_bank; scope transactions:write. Registry, scope map, load-routes, spec snapshot and the generated accounted-api skill updated; five BANK_SYNC_* / BANK_SESSION_EXPIRED codes added to the structured-error registry. The web Synka-nu route is left as is (see DECISIONS.md). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QdcDV7CngLkWUvfKSsxFhS * test(bank-sync): use the options object in the remaining chip-state calls Four multi-line calls still passed the clock positionally after getChipState moved to an options object; tsc flagged them (vitest did not, the extra argument was ignored at runtime). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QdcDV7CngLkWUvfKSsxFhS * fix(api): address skeptic findings on the agent-triggered bank sync Three refutations from the pre-publish skeptic pass: 1. Core imported the extension. The v1 sync route pulled the runner straight from @/extensions, which the core-build gate rejects and which left a live bank endpoint on zero-extension builds. The route now resolves it through the registry's services channel against a contract in lib/bank-sync/trigger-sync-contract.ts (same pattern as the Skatteverket read service) and answers EXTENSION_DISABLED when the extension is absent. 2. The idempotency cache stored the handler-level 429. A same-key retry after Retry-After, which is the documented retry, replayed the stale cooldown as a 400 for the cache's 24-hour TTL. withApiV1 no longer caches 429 responses; regression test added. The endpoint's pitfall no longer claims Idempotency-Key is mandatory (it was never enforced). 3. Two cron tests read the clock twice and failed whenever a millisecond passed between the reads. They now pin the clock with fake timers. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QdcDV7CngLkWUvfKSsxFhS * fix(bank-sync): durable cooldown lease and review wording Resolves the PR #2165 review findings in one pass. Superagent P1: the attempt throttle was a process-local Map, so two agent calls on different serverless instances (or a retry after a cold start on a failing connection) could each bill an Enable Banking call, contradicting the one-sync-per-15-minutes promise. New bank_connections.sync_lease_until (migration 20260902150000), claimed with one conditional UPDATE before the bank is called; Postgres row locking makes exactly one claimer win, the rest answer BANK_SYNC_COOLDOWN. The lease stays for the full window on success and failure. Tests cover the claim order, a failed attempt seen from a second instance, a lost race, and an expired lease. CodeRabbit: the =1 plural branch now reads "in 1 day" / "om 1 dag" (daysUntilConsentExpiry rounds a partial day up, so "tomorrow" could be today); the cooldown pitfall on the v1 endpoint, the MCP description and the in-band cooldown instruction now say a cooldown can follow a failed attempt and tell the agent to compare last_synced_at before deciding. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0125TMQQBjBBZG9YxP7wQWub * fix(bank-sync): lease claim as a literal filter for the schema guard CI's no-phantom-columns guard counts runtime-built query expressions and its ceiling is exact; the templated `.or('sync_lease_until.is.null,...')` claim added one. The column now defaults to epoch (NOT NULL), so "never claimed" is just "expired long ago" and the atomic claim is a single literal `.lte('sync_lease_until', now)` the guard can check. Migration is unshipped (same PR), so it is edited in place. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0125TMQQBjBBZG9YxP7wQWub * fix(bank-sync): runner verifies company membership before the lease Superagent (round 3): the MCP path reached the shared runner without a membership check of its own. Both callers do enforce it upstream (withApiV1's company resolution and resolveMcpCompanyContext in the MCP dispatcher), but the runner writes transactions and bills a bank call, so it now checks company_members itself, before the cooldown and the lease claim, and answers NOT_FOUND for a non-member. The viewer check that was buried inside the sync block moves up with it. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0125TMQQBjBBZG9YxP7wQWub --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
a80ce54b78 |
fix(mcp): eager-auth flag on the Grok connector links so Grok starts OAuth (#2167)
* fix(mcp): eager-auth flag on the Grok connector links so Grok starts OAuth Live test after #2158: pasting the Grok URL into grok.com's custom connector dialog listed all 150+ tools and never opened the sign-in. Grok probes the URL without credentials, like claude.ai, and reads the lazy 200 on initialize as an authless server; only the 401 challenge starts OAuth (#2159 fixed the same thing for the claude.ai link). - lib/onboarding/checklist.ts: mcpServerUrl() builds the server URL with an optional eagerAuth flag; sideDoorServerUrl() gives the Grok side door auth=required and keeps ChatGPT lazy; claudeConnectorLink() reuses it. SIDE_DOORS / SideDoor move here from the component. Tests for all three. - NewUserChecklist copies the door-specific URL (now with a client marker). - ApiKeysPanel's Grok row copies the flagged URL, mirroring the Claude one. - auth-mode.ts comment records the second consumer; registry entry's Grok step carries the flag; DECISIONS. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LhTJcwgzmN3TsLR8tVHwdi Signed-off-by: Emil <emilmattsson14@gmail.com> * docs(mcp): registry Claude.ai step carries auth=required too Review pass on #2167: the registry entry flagged the Grok install URL but left the Claude.ai step on the bare URL, which pre-fills "None" in claude.ai's dialog (#2159). Same file, same flag, now consistent. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LhTJcwgzmN3TsLR8tVHwdi Signed-off-by: Emil <emilmattsson14@gmail.com> --------- Signed-off-by: Emil <emilmattsson14@gmail.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
678acfe7ef |
fix(mcp): eager-auth flag so claude.ai's connector dialog detects OAuth, not "None" (#2159)
claude.ai's two-step "Add custom connector" dialog probes the server URL without credentials and pre-fills the Authentication choice from the answer. Our lazy-auth endpoint (issue #1814) answers 200 on an anonymous initialize, which the dialog reads as an authless server: it suggests "None", and a connector added with that default never opens the sign-in when the challenge arrives later. Per Anthropic's connector docs a 401 is the only answer it reads as OAuth ("Claude does not honor a WWW-Authenticate header on a 200 response"). - `auth=required` on the endpoint URL (extensions/general/mcp-server/ auth-mode.ts) turns lazy auth off for that URL: every tokenless request, initialize included, answers the 401 + WWW-Authenticate challenge. Callers with a token are unaffected; the bare URL keeps lazy auth for Claude Code, the plugin, Cursor and ChatGPT, and existing connector records are untouched. - The links we control carry the flag: Settings -> API & MCP (install link and copy block), the onboarding checklist, both docs pages and claude-plugin/CONNECTORS.md (plugin 1.2.3). The docs' Path A now describes the eager flow (sign-in opens on Add) instead of telling users to override the dialog's "None". - Tests: eager-auth.test.ts (401 on initialize/tools/list/public tools, namespaced metadata pointer, token no-op, exact-flag only); checklist link shape updated. Companion: gnubok-website PR (Kom igång connector link + regenerated connect-claude / anslut-claude pages). Claude-Session: https://claude.ai/code/session_013yw62FMXGSzo6icFDiBwP3 Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
f266c386f3 |
chore: repo-wide bloat sweep, remove dead code and fold duplicate helpers (#2150)
* chore: repo-wide bloat sweep, remove dead code and fold duplicate helpers Remove 33 dead files, ~270 unreferenced exports/types, 13 dead i18n namespaces and 4 unused dependencies; fold byte-identical helper copies into one canonical home each (lib/utils chunk/sleep/utcDateStamp, lib/dates/iso, lib/invariants/uuid, lib/xml/escape, lib/reports/sru/format, lib/pdf/number-text, lib/browser/panel-request, lib/api/v1/body + v1ValidationError rolled out to ~55 v1 routes, booking-template schemas). No behaviour change: v1 bodies and status codes, MCP tool schemas, DB writes and money math are untouched. Naive ore rounding was deliberately not swapped for roundOre; see DECISIONS.md 2026-09-02 for the full list of things left alone on purpose. tsc, lint, 19588 unit tests and check:guards green; antipattern baseline ratcheted (naive-ore-round 622 -> 620, hand-rolled-invariant 115 -> 113). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * test(transactions): import RawTransaction from @/types after the ingest re-export removal CI's type ratchet (check:types, full tsconfig) caught the one test file that still imported the type through lib/transactions/ingest. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
18cbc4c30a |
fix(security): audit remediation 2026-09-01: api_keys identity, viewer gates, OAuth binding, XSS, MFA gate (#2155)
* fix(security): bind api_keys to the caller, lock hash-as-bearer RPCs and provider token tables Security audit 2026-09-01, critical items. - api_keys INSERT requires user_id = auth.uid() again (an admin could forge a key for any co-member and act as them in every company they belong to); SELECT is own-keys-or-admin; a BEFORE trigger freezes the identity and credential columns against user-session UPDATEs. - rotate_mcp_refresh_token and validate_and_increment_api_key become service_role only: they match rows by a presented SHA-256, so a hash readable by co-members was a bearer credential. - validate_and_increment_api_key fails closed when the key's user is no longer a member of the key's company. - provider_consent_tokens and provider_otc: the DELETE policies collapsed to "caller has any team row" (correlated subquery on a non-existent team_members.company_id). All member policies dropped; service_role only, matching every existing code path. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): role gates, ownership guards and posting integrity in the database Security audit 2026-09-01, high items at the database layer. - One table-level guard, enforce_company_writer_role(), blocks the read-only viewer role on 55 company-scoped tables including through the 15 membership-only SECURITY DEFINER writers. Keyed on the JWT role claim so it fires inside definer bodies; no-op for service_role and trigger cascades. - company_members user_id/company_id immutable from user sessions; invitations can never grant owner; team_members gains a transition guard (admins keep non-owner role moves); companies team_id and archiving are owner-only and team attachment needs team membership. - Direct statements (current_user = authenticated) can no longer insert posted headers, add lines under posted verifikat, or post a draft with a voucher number the sequence never issued. Sanctioned RPCs run as the definer and are untouched; the engine's own draft-then-post shapes still pass. - create_document_version refuses viewers and foreign storage paths; validate_version_chain needs membership and loses anon EXECUTE; match_documents / match_booking_templates lose anon; cron maintenance RPCs become service_role only; the production-only seed_asset_categories is dropped. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * build: pin tsx as an exact devDependency instead of fetching it with npx at build time prebuild ran "npx tsx" with no lockfile entry, so every Vercel, Docker and CI build downloaded tsx@latest and its transitive tree from the registry with no integrity check, inside the build environment. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): refuse the viewer role on API-key and MCP write paths The v1 wrapper and the MCP company routing checked company membership but never role, and both run as service role, so a read-only viewer holding an API key could post vouchers and change settings through the API. Mutating methods and non-read scopes now return 403 ROLE_READ_ONLY for viewers on v1; MCP write tools refuse viewers the same way. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): stop serving uploaded SVG, XML and HTML as executable content on the app origin Uploads persisted the browser-declared mime type and the inline proxy served it verbatim, sandboxing only text/html; the storage proxy forwarded the uploader's Content-Type. Any writer, or any Peppol sender, could plant a scripted SVG or XHTML that executed on app.gnubok.se. - inline route: allow-list of natively safe types (PDF, raster images) served as before; everything else gets the opaque sandbox CSP. - storage proxy: octet-stream + attachment + sandbox unless the DB mime for the key is on the allow-list. - document-service: the stored mime is the magic-byte validated type. - logo upload: magic-byte validation, SVG refused. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): byrå brand logo upload decides the type by magic bytes and drops SVG Same pattern as the company logo route: the logos bucket is public, so a scripted SVG (or anything declared as an image) must never land there. The upload pickers stop advertising SVG. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): bind Enable Banking, Stripe and WooCommerce callbacks to the initiating user The callbacks resolved the pending row by oauth_state alone, so a victim who completed an attacker-initiated consent had their bank account, merchant account or store attached to the attacker's company. requireFlowInitiator() now requires the cookie session of the user who started the flow: no session redirects to login with the callback URL preserved, a different user is refused and nothing is exchanged. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): guard tenant-controlled outbound fetches and surface the disabled rate limiter WooCommerce and Shopify syncs fetched a member-editable store URL with plain fetch() and redirect following under the service role, and the invoice PDF renderer fetched company_settings.logo_url unguarded. All three go through a new safeFetch() (public-IP validation via url-guard, https only, redirect: 'manual', body size cap) and re-normalise the stored host at use time. checkRateLimit() keeps failing open on hosted but logs one error per process when Upstash is not configured and exports isRateLimiterConfigured(). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): decide the API MFA gate from server-authenticated factors, not the session cookie getAuthenticatorAssuranceLevel() without arguments derives nextLevel from session.user.factors, which comes from the unsigned sb-*-auth-token cookie. Deleting factors from the cookie made an enrolled account look like it had nothing to step up to, on every /api route and in requireAuth. Both gates now read factors from the getUser() result or listFactors() and the level from the verified JWT claim, and fail closed on errors. Page-branch gate hardened the same way. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): bind Fortnox/Visma, Gmail and Skatteverket callbacks to the initiating user The arcim-migration callback exchanged the provider code onto whatever consent the one-time state named, with no check of who completed the flow and no org-number comparison, so a phished Fortnox admin handed their ledger to the attacker's company. provider_otc now records the initiating user (migration 20260902100000); the callback requires that session and, after the exchange, refuses a provider company whose org number differs from the consent's company. The Gmail and Skatteverket callbacks enforce the same initiator check. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): BankID signup confirms the email before linking the identity Signup created an email-confirmed, MFA-exempt account for any address the caller typed and returned a magic link, so an attacker could pre-register a victim's email and keep a permanent BankID login into the account the victim later adopted. The user is now created unconfirmed, the identity carries email_verified_at NULL (migration 20260902101000), bankid_linked is not set until the mailed confirmation is clicked, and BankID login of a pending identity is refused with the confirmation re-sent. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(security): bind MCP OAuth redirect URIs to the consenting user and cap scopes A user-registered redirect URI was allowlisted globally, the consent page named no client, and all scopes were pre-checked, so one phishing link handed an attacker a full-scope key for the victim's company. Registered URIs now resolve only for the registrant or a colleague sharing a company; the consent page shows the client identity and redirect host; non-built-in clients default to read-only pre-checks; scopes are capped by the user's role (viewer: read only) at consent and at /token. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(auth): client follow-ups for BankID confirmation, callback mismatch copy and decision log - register client handles the new confirmation_sent response from BankID signup with the existing inbox screen instead of calling verifyOtp. - BankID login surfaces the email_unconfirmed explanation. - WooCommerce settings map woocommerce_error=wrong_user to its own copy. - Logo help text no longer advertises SVG. - DECISIONS.md records the audit remediation choices. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * fix(mcp-oauth): literal SoD columns in the api_keys insert so the phantom-column scanner resolves them Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * test(logo): type the upload fixtures as Uint8Array<ArrayBuffer> so they are valid BlobParts Fixes the typecheck ratchet on PR #2155 and ratchets the baseline down by the one legacy error the change removed. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
6e8d76a9cb |
fix(skattekonto): remove the drift email, its event and the unused drift route (#2149)
The nightly skattekonto sync emailed "Skattekontot stämmer inte med bokföringen" whenever Skatteverket's saldo differed from BAS 1630 by more than 1 kr, every 24 hours while it lasted. On 2026-09-02 it fired on a 35 842 kr gap that the reconciliation explained to the last krona with 14 unbooked rows, while the Hem notice and the reconciliation page (both gated on unexplained_difference) said nothing was wrong. The check shipped in May 2026 (#525) before any in-app skattekonto view existed; the dashboard tile its comments promise was never built and the drift API route had no consumer. Since 2026-08-25 the reconciliation page and the Hem notice are the surface, with one definition of "stämmer inte". Removed: skattekonto-drift.ts, skattekonto-drift-email.ts, their tests, the skattekonto.drift_detected event type, the handler registration, the cron's drift hook, GET /api/extensions/skatteverket/skattekonto/drift, and the ROPA activity for the mail. The route is dropped from the ungated extension route allowlist to lock the ratchet. skattekonto_drift_tolerance stays: the Hem notice reads it. Stale skattekonto_drift_last_alert_at rows in extension_data are inert. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
867767a22f |
feat(inbox): document-type badge and filter, +lev/+ver plus-addressing (#2129) (#2148)
* feat(inbox): document-type badge and filter, +lev/+ver plus-addressing (#2129) Phase 1: every inbox row shows its document kind (Kvitto, Leverantorsfaktura, Myndighetsbrev, Ovrigt) from the existing AI documentKind, and a second menu next to the status filter narrows the list to leverantorsfakturor or underlag. Pure predicate in lib/documents/inbox-kind.ts with tests. Phase 2: the shared inbox address accepts RFC 5233 plus-addressing. The webhook splits the local part at the first + and looks up the base, so <local>+anything@ now reaches the company instead of 404ing. +lev and +ver land in the new nullable invoice_inbox_items.kind_hint column (CHECK supplier_invoice | receipt), threaded through EmailMeta into both inbox inserts and returned by GET /items. kind_hint wins over documentKind for the badge and the filter and survives re-extraction because it is a column. The sources panel shows both tagged addresses with a one-line hint (sv + en). Tests: filter predicate per kind and null; parser and tag mapping; webhook routes +LEV and an unknown tag; pg test pins the CHECK and NULL default. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Hzv2Z2eCq8iJAAe8XC1hNr * fix(inbox): honest empty state under a type filter, detail pane shares the row's kind resolution Skeptic findings on #2148: with a type filter narrowing 'Att göra' to zero the empty state claimed 'allt är bearbetat' while the status trigger still counted pending rows; it now says no items of that type are here (sv + en). The fields rail printed the AI documentKind only, so a +lev hint could disagree with the row badge; it now uses resolveInboxKind like the list. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Hzv2Z2eCq8iJAAe8XC1hNr * fix(inbox): keep the type-filter empty state off purchase lists, carry kind_hint onto rejected attachment rows CodeRabbit on #2148: the purchase lists (Saknar underlag, Hämta från portal) ignore the type menu, so a leftover kind filter must not pick their empty-state copy. A rejected attachment (unsupported MIME, too large) now keeps the sender's +lev / +ver hint on its error row like every other inbox insert; the allowlist test covers it. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Hzv2Z2eCq8iJAAe8XC1hNr * fix(inbox): set the +lev/+ver kind hint only when the shared address resolved the company CodeRabbit on #2148: the hint was computed before recipient resolution, so a tag on an unknown or retired shared address could ride along onto a custom-domain match. It is now assigned inside the active shared-inbox branch only; regression test covers the multi-recipient case. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Hzv2Z2eCq8iJAAe8XC1hNr --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Mattsson
|
8b09b06e14 |
feat(skatteverket): ombudsregister grant verification, honest session expiry, daily ombud sync (#2130)
* feat(skatteverket): ombudsregister grant verification, honest session expiry, daily ombud sync
Users reported the Skatteverket connection "just disappearing" with no
banner, needing BankID again every time. Two causes, both fixed here:
1. SKV's per-flow refresh token lives 65 minutes. /status and the
skv_disconnected notice called any stored refresh token "refreshable",
so a days-dead session reported healthy and the reconnect banner never
fired until a submission failed live. lib/skatteverket/session-lifetime
now decides refreshability (expires_at + 5 min, refresh cap) for both
surfaces; the settings panel states the one-hour session lifetime.
2. The durable fix is the ombud (system certificate) path, dormant since
July behind SKATTEVERKET_SYSTEM_AUTH_MODE. Skatteverket added scope
`obr` (Ombudshantering v2) to our application id on 2026-09-01, so grant
verification can now ask the ombudsregister instead of classifying 403s
from the read services:
- lib/ombud-client.ts: GET /ombud/autentisieratOmbud, GET /roller,
POST .../djuplank/utseombud, on the system identity, per the public
tjanstebeskrivning v2.0 (mirrored in dev_docs/skatteverket/ombudshantering).
Role codes are env-pinned (SKATTEVERKET_OMBUD_ROLL_LASOMBUD/_MOMS) or
matched on rollbeskrivning text; a deep link never mints with a
guessed code.
- grant-probe.ts: register first, read-service probes only as fallback.
- New daily cron /api/extensions/skatteverket/ombud/sync/cron (30 3 * * *):
one register call discovers every company that granted us, creates or
downgrades connection rows by org number, runs from shadow mode on,
and never mass-revokes on an empty register.
- POST /system-connection/deeplink + "Utse {app} som ombud" button:
the company lands in SKV's e-service with roles pre-selected.
- Default system scopes include `obr`; skvRequestWithAuth gains an
`accept` option (Ombudshantering requires the Accept header).
Still inert in prod until the org certificate and avtal land; the cron and
verify routes no-op while system auth is off or unconfigured.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HkLnhWnxt5wWB9j3vfMmxu
* fix(skatteverket): skeptic round on ombud sync, register 404 fallback, opt-in-only rows, mass-downgrade guard
Cron touches only existing connection rows (a tenant's own Verifiera or
deep-link opt-in; the deeplink route now records a pending row), so an
org-number twin never gets auto-verified, and rows the tenant revoked
locally stay revoked. A register 404 throws by default (spec: wrong URI)
and is empty only for the cron, which guards it. Decisions are planned
before any upsert; a run that would fully deny >= 3 rows and > 50% of the
granted ones applies no downgrade. Grants that classify as neither
behörighet are 'error', not 'denied'. Literal select in listConnections for
the phantom-column scanner. window.open without 'noopener' so the
pre-opened tab exists; opener nulled by hand. Deeplink route test added.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HkLnhWnxt5wWB9j3vfMmxu
* fix(skatteverket): CodeRabbit round: exact role labels, paginate connections, deny never-listed rows, fail deeplink without opt-in row
Role descriptions match the whole label so 'Momsdeklaration,
deklarationsombud' is never read as the narrow moms role. listConnections
pages through fetchAllRows on (created_at, id). A pending row the register
never lists is written once as denied instead of staying 'Inte verifierad'.
The deeplink route returns 500 when the opt-in row cannot be stored, and the
panel navigates in-tab when the pre-opened tab was blocked.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HkLnhWnxt5wWB9j3vfMmxu
* fix(skatteverket): fence ombud grants on contested org numbers; cron honours unrecognised role codes
An org number claimed by more than one live company is contested: verify
and deep link answer 409 ORG_NUMBER_CONTESTED and the nightly sync changes
nothing on it, so a tenant that typed a victim's public org number cannot
inherit the victim's grant. The sync also skips huvudmän whose register
roles classify as neither behörighet (pinning problem, never a denial),
mirroring probeViaOmbudsregister.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HkLnhWnxt5wWB9j3vfMmxu
* fix(skatteverket): validate the ombud deep link host; withdraw grants on contested org numbers
The register's djuplank must be an https skatteverket.se URL before it is
returned or navigated to (the settings page follows it). The nightly sync
now withdraws a grant already recorded on an org number that more than one
live company claims, instead of only refusing new ones.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HkLnhWnxt5wWB9j3vfMmxu
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
|
||
|
Mattsson
|
d8cf78330e |
docs(self-hosting): own-credentials section, stale connector lines, complete .env.example (#2146)
* docs(self-hosting): own-credentials section, stale connector lines, complete .env.example (#2131) SELF-HOSTING.md said the Skatteverket client wiring "ships in a following release"; PR #2103 merged it, so both bank sync and Skatteverket now carry traffic through the hosted proxy with a key. The two stale sentences are replaced and SOVEREIGN.md line 48 says the same thing. New "Own credentials (no connector key)" subsection documents the path an operator takes without a key: Enable Banking app in restricted production mode with the callback URL, the Skatteverket developer-portal application with the redirect URI, every variable the code reads, the five production base URLs (all defaults point at the test environment), the kill switch, and the rule that any own credential switches that upstream out of connector mode. .env.example gains the Skatteverket block, the optional Enable Banking variables, and RESEND_INBOUND_DOMAIN / RESEND_INBOUND_WEBHOOK_SECRET, which the invoice-inbox manifest requires but the example never listed. DOCKER.md no longer claims Enable Banking is excluded from the self-host preset (docker/extensions.self-hosted.json ships it). ENABLE_BANKING_SANDBOX is removed from the enable-banking manifest and the index.ts header: declared as optional, never read anywhere; the sandbox is selected by ENABLE_BANKING_API_URL. Logged in DECISIONS.md. Closes #2131 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012C6M2ZoZc6QDRxU3m9WzgE Signed-off-by: Emil <emilmattsson14@gmail.com> * docs(self-hosting): correct key format, AISP scope caveat, SKV scopes and rotation note (#2131) Skeptic findings on PR #2146, one pass: - ENABLE_BANKING_PRIVATE_KEY: the decoder base64-decodes first and wraps anything else as DER, so a raw PEM fails at JWT signing. The docs and .env.example no longer claim it is accepted. - Enable Banking restricted mode covers the operator's own accounts only; an instance hosting client companies is doing licensed AIS and needs the connector key or its own AISP registration. Said so. - Listed the OAuth scopes the app requests (both AGI scopes), noted that the kill switch gates API calls, not the BankID login, and that the token encryption key has no dual-key rotation. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012C6M2ZoZc6QDRxU3m9WzgE Signed-off-by: Emil <emilmattsson14@gmail.com> --------- Signed-off-by: Emil <emilmattsson14@gmail.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
dd84d6c1bb |
fix(mcp): tag unmapped tool failures with their cause vocabulary (#2051) (#2135)
Closes #2051. errorCauseTag() shipped in #2027 written and tested but wired to nothing. This connects it: the two execution catch paths (sync call and task) now pass errorCause into mcp.tool_called, carrying the SQLSTATE or coded-error code, else the error's class name, capped at 64 chars. The rows this exists for are the UNKNOWN_ERROR residue, whose errorMessage is the constant "Något gick fel. Försök igen." and whose errorDetail is the English constant: 465 such rows in the last 30 days (create_voucher 58 of its 60 failures, query_journal 122) with nothing to cluster on. A five-character SQLSTATE is protocol vocabulary; a raw driver message can quote row values from a constraint violation and belongs in the server log, never in event_log, so the raw message is deliberately not captured. A plain `new Error(...)` tags null rather than 'Error': tagging everything is the same as tagging nothing. Pre-execution denials (scope, capability, validation, unknown tool) pass nothing because their errorCode already is the cause. Self-tested by unwiring one call site and watching the new test name it. Claude-Session: https://claude.ai/code/session_01L3P2hr19PhQuCoTSGoegcY Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
53c9c0c4f9 |
feat(mcp): second examples batch: five more tools, seven examples, 80 tokens (#2137)
Continues #2100 under the margin the envelope trim (#2123) created: 58 999 to 59 079 against the 60 000 ceiling, spending under half the room and leaving ~920. Picked by evidence, not traffic alone: gnubok_search_tools was sent a nonexistent `offset` in prod today, so its examples show the actual levers (query, detail, limit) and the description of the mistake; the create/complete document-upload pair's examples ARE the two-step flow, same upload_id and file_name on both sides; list_uncategorized_transactions is the highest-traffic read (15 122 calls/30d) and gets the pagination shape; link_document_to_voucher gets the minimal linking call. All seven pass the input-examples validation suite (same unknown-key guard the server runs, plus required/type/enum/pattern and the placeholder-id check), and the pinned tool list is updated. Claude-Session: https://claude.ai/code/session_01L3P2hr19PhQuCoTSGoegcY Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
8d9bf383d1 |
perf(mcp): trim the one schema 58 catalogued tools share, reclaiming 2 552 tokens (#2123)
The catalog had 116 tokens of headroom and server.ts took 70 commits in the preceding 14 days, so the next ordinary tool addition would have failed CI. Measured before cutting, which is what made this findable: outputSchema is 38% of the whole catalog (23 290 tokens), and STAGED_OPERATION_SCHEMA alone accounts for 14 736 of it, the same envelope transmitted 58 times. Descriptions, which the three previous rounds trimmed, are 10%. Three edits to one constant, no tool demoted and no field removed: period_status stops declaring its three sub-properties as JSON Schema and carries them in one sentence (actor, approve and preview were already bare objects, so this makes the envelope internally consistent), the next hint drops args' redundant additionalProperties: true, and operation_id's description loses six words. Every change is in the looser direction on purpose. The server emits structuredContent for every tool and the documented failure mode is a declaration too TIGHT making a strict client reject a successful call; a looser one cannot do that. next keeps additionalProperties: false because staging.test.ts pins it as a closed shape, and a guard whose reason is not in front of you is not a guard to loosen for 420 tokens. Ceiling ratcheted 61 600 to 60 000, leaving ~1 070 tokens of deliberate working margin rather than the ~300 the previous rounds left. The bench log records the cycle that margin causes: ratchet tight, block the next feature, bump, demote. If more is needed the lever is priced: the envelope still costs ~11 800 tokens across those 58 tools, and the fix is to stop repeating it, not to trim it further. Claude-Session: https://claude.ai/code/session_01L3P2hr19PhQuCoTSGoegcY Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Mattsson
|
b56da5d6c5 |
feat(api): expose bank-connection freshness in MCP and v1 REST (#2124)
* feat(api): expose bank-connection freshness in MCP and v1 REST
gnubok_connect_bank now returns last_synced_at, consent_expires and
error_message per connection, and its instructions tell the agent to
flag stale or expiring connections. New read-only endpoint
GET /api/v1/companies/{companyId}/bank-connections exposes the same
fields to API-key integrations (scope companies:read).
Background: a user's PSD2 feed died silently in July; bookkeeping
looked complete while three weeks stale, and nothing on the API/MCP
surface could reveal it. Sync stays cron-driven; an agent-triggerable
sync was considered and deferred (see DECISIONS.md).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01USJHnxsindrs9X6zqQDLix
* fix(api): address skeptic findings on bank-connection freshness
- Map the bank-connections group into skills/accounted-api (apiskill:check
crashed on the unmapped group; regenerated skill files included).
- Gate the v1 route on the bank_sync capability, mirroring the MCP twin:
a lapsed entitlement now answers with a capability error instead of
status=active with a frozen last_synced_at.
- Reword MCP instructions + v1 pitfalls: null last_synced_at right after
connecting is normal, staleness threshold aligned to the UI's 36 hours,
and re-authorisation is only advised for expired/error/consent-out, not
for stale-but-active connections (lapsed subscription or deselected
accounts are the usual causes there).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01USJHnxsindrs9X6zqQDLix
* fix(mcp): keep gnubok_connect_bank schema under the tools/list token ceiling
The enriched outputSchema plus the worked examples that landed on main
(#2100) pushed the projected tools/list payload 20 tokens over the
61.6K context-budget ceiling. Drop the per-property descriptions from
the new freshness fields; the instructions string (runtime output, not
catalog payload) already explains them.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01USJHnxsindrs9X6zqQDLix
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
169e7eaf4e |
feat(mcp): worked examples on five high-traffic tools, and in the error that rejects a call (#2100)
#2066 asked for input_examples on the top ~20 tools by call volume. The binding constraint turned out to be budget, not writing: after #2089 reclaimed 3 763 tokens, its own policy required ratcheting the tools/list ceiling down with it, so the real headroom was 317 tokens. Ten examples across five tools cost 199, leaving ~118. The ceiling is not raised. Tools picked from 30 days of mcp.tool_called crossed with the combinations the descriptions already warn about and callers still get wrong: account_override without an explicit vat_treatment (books gross, no moms line), representation without deltagare and syfte, confirmed on a high-risk approval, a balanced voucher where the moms leg is its own line, and get_kpi_report, where one caller sent `metric` 604 times over seven days to a tool whose only parameter is period_id. Examples are also surfaced in the unknown-parameter error. That costs nothing in tools/list, because it only ships on the response to a call that already failed, and it reaches the caller that most needs it: a key list told DueCue's agent which parameter was wrong but not what a correct call looks like, and the same rejected call repeated for a week. Every example is validated against its own schema by the same findUnknownArgKeys guard the server runs, plus required/type/enum/pattern checks. An example our own boundary would reject is worse than none: it teaches the exact mistake the guard then punishes. That test caught three invented enum values in this change's own first draft. Claude-Session: https://claude.ai/code/session_01L3P2hr19PhQuCoTSGoegcY Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Mattsson
|
cd40127f0e |
feat(bank): expose bank-reported balance (booked + available) in UI, reconciliation, MCP and v1 API (#2118)
* feat(bank): expose bank-reported balance (booked + available) in UI, reconciliation, MCP and v1 API
The PSD2 sync has fetched the bank's reported balance for years but the
data was stranded (F7): the Bank-page source picker read a cash_accounts
column no sync ever updated (frozen at connect time), reconciliation
hard-coded external_balance to null for bank accounts, and neither MCP
nor the v1 API exposed any balance at all, so the only path to a current
bank balance was logging into the bank.
- getAccountBalance now returns booked + available from the same
quota-limited BALANCES response (previously all but one type discarded)
- every sync (manual + cron) mirrors balance, available_balance and
balance_updated_at into cash_accounts, fixing the stale picker
- new cash_accounts.available_balance column (additive migration)
- reconciliation bank kind: external_balance = bank-reported balance,
plus bank_reported_* fields and fetch timestamp in the bank block;
difference math stays movement-based and untouched
- reconciliation view shows "Saldo enligt banken ... hamtat {date}"
- MCP gnubok_list_cash_accounts returns the three balance fields; the
cash_today prompt now reports the bank's figure instead of teaching
agents to answer with the bookkept 19xx balance
- new GET /api/v1/companies/{companyId}/cash-accounts endpoint
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ewu46quXgh9LSr9UwYusxm
* fix(bank): keep external_balance null for bank sign-offs; never fabricate a zero balance; guard the mirror against stale writers
Post-review fixes from the skeptic pass + CodeRabbit on PR #2118:
- external_balance stays null for the bank reconciliation kind: sign-off
persists it into account_reconciliations and bokslutsbilagor computes
closing - external from that row, so a today-balance stored on a
balansdag sign-off printed a phantom warning-red differens in the
year-end appendix. The bank-reported figure lives only in the
timestamped bank_reported_* pair in the bank block, and only when its
fetch timestamp exists (a balance of unknown age is suppressed).
- AccountOverview no longer falls back to today's date when the balance
timestamp is missing; the line is omitted instead.
- getAccountBalance returns null on an empty BALANCES response instead
of fabricating amount 0 with a fresh timestamp; sync keeps the
previous stored value.
- updateBalancesFromSync only writes over an older-or-missing
balance_updated_at, so an older sync run finishing later cannot move
the mirrored balance backwards.
- The inline initial backfill (picker save) now mirrors fetched
balances into cash_accounts too (accounts_data is deliberately not
re-written there).
- cash_today MCP prompt mentions the gnubok_call_tool bridge for hosts
that only see the default catalog.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ewu46quXgh9LSr9UwYusxm
* fix(bank): express the stale-writer guard as two literal predicates for the schema guard
The .or() with a template literal pushed the no-phantom-columns
unresolvable-expression count over its ceiling. Same semantics, two
updates: one for rows with an older timestamp, one for rows with none.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ewu46quXgh9LSr9UwYusxm
* fix(bank): rank interimBooked (ITBD) as a booked balance type before the generic fallback
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ewu46quXgh9LSr9UwYusxm
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
||
|
Mattsson
|
a57a8d968b |
fix(webshop-orders): stop syncing failed WooCommerce orders, remove stale unpaid rows (#2119)
* fix(webshop-orders): stop importing failed WooCommerce orders, remove stale rows on failed transition Failed checkouts carry no money event but imported as permanently unbookable 'Ej betald' rows (user report). orderImports() now excludes 'failed' alongside 'trash', and a re-polled order that transitioned to failed deletes its existing row via removeWebshopOrders(), which enforces the freeze boundary app-side: frozen rows are never deleted, and a parent with a frozen refund child is spared because parent_order_id cascades. Removal failures hold the sync cursor like upsert failures do. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014hQZLCdT56j8nAyoQtAs2C * fix(webshop-orders): make failed-order removal race-safe per skeptic findings Repeat every guard on the DELETE statement itself, not only the candidate select: a row booked/marked/invoiced between the two round trips must survive (TOCTOU refutation). Never remove paid rows (orderRemoves gated on !orderIsPaid plus is_paid=false on both statements): money moved at some point, and paid parents are the only rows that can carry refund children, which also closes the cascade race without a DB trigger. Spare cross-marked rows (legacy_transaction_id): the order may be booked via the retired transactions feed without any freeze column set. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014hQZLCdT56j8nAyoQtAs2C * fix(webshop-orders): audit-log successful failed-order removals Compliance swarm finding (ISO A.8.10): the hard delete logged only its failure path. Every successful removal batch now logs companyId, deleted row ids and the requested external_ids, the only deletion record for pre-bokforing rows that carry no behandlingshistorik. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014hQZLCdT56j8nAyoQtAs2C --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
fa69174aa0 |
fix(bank): never pre-check or mirror another company's accounts in the EB callback (#2116)
* fix(bank): never pre-check or mirror another company's accounts in the EB callback At one-session banks (SEB) the PSU's single consent can cover accounts a sibling company books. The OAuth callback stored whatever the session returned into the active company: all pre-enabled, mirrored into its cash_accounts, ledgers allocated from its chart: one 'Spara val' away from booking another aktiebolag's transactions (user report F1, 2026-09-01). The deliberate reuse path (findReusableSessions) already guards claimed IBANs; the callback now runs the same check via fetchCrossCompanyAccountContext: - accounts claimed by another of the user's companies are stored disabled + flagged (claimed_by_company_*), skipped by the cash_accounts mirror, and the picker names the claiming company - a 'Synkas ej' deselection made on any other connection row is carried onto fresh rows (the recurring came-back-pre-checked complaint, C2) - lookup failure fails closed: new accounts stored deselected - accounts the row itself already carried keep their own state, so a renewal can never switch a working feed off Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0197wmwP6zNaYvsGfbZuQHGA * fix(bank): close the skeptic-found holes in the cross-company claim guard Consolidated fixes from the three-skeptic review of PR #2116 (all three refuted the first cut): - Active-company standing state (enabled cash_accounts + enabled accounts on its live-ish connection rows) now outranks sibling claims company-wide, not row-wide: a bank-list renewal arrives on a FRESH row with no priors, and the old row-local check would have let a sibling claim switch a working feed off while supersede demoted its cash row. - pending_selection rows no longer claim accounts or feed deselection memory: their flags are unconfirmed callback output (including this guard's own fail-closed writes), so an abandoned picker or a transient lookup error can no longer poison later connects. - Guard-disabled accounts are never mirrored from the callback: upsertFromPsd2 with enabled:false for a new-to-row account could promote the seeded primary 1930 manual row and flip it to disabled under a foreign identity. - The selection save skips ledger allocation and the cash_accounts mirror for disabled never-mirrored accounts, so 'no cash row, no 19xx slot burned' holds past the mandatory Spara val, and strips the claimed_by_*/deselected flags when the user deliberately enables an account. - Deselection carry is no longer silent: deselected_elsewhere flag + picker note 'Tidigare bortvald'. - Claim lookups paginate via fetchAllRows: the bare select's silent 1000-row PostgREST cap failed open for exactly the multi-company consultants the guard exists for. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0197wmwP6zNaYvsGfbZuQHGA * fix(bank): claim-guard round 2: pending_selection claims asymmetrically, paged reads ordered Skeptic re-verification of 25a339810 found two holes: - Excluding pending_selection rows from claims reopened the attach-to-picker window: an attach-created row holds deliberately offered enabled accounts with no cash_accounts rows until its picker is saved, and a full-OAuth connect in another company inside that window could take the same physical account. Enabled accounts on pending_selection rows claim again; their disabled flags still stay out of the deselection memory (unconfirmed callback output, including the guard's own fail-closed writes). - Both fetchAllRows claim queries now order('id'): unordered .range() pagination can silently skip rows at page boundaries, and a skipped row is a missed claim, failing open at exactly the 1000+-row scale the pagination was added for. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0197wmwP6zNaYvsGfbZuQHGA --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
f1d76deaba |
fix(providers): stop dead-ending on a resource 403, and stop dropping every migrated kreditfaktura (#2113)
* fix(providers): stop dead-ending on a resource 403, and stop dropping every migrated kreditfaktura
Two independent defects in the provider migration, both customer-visible.
A per-resource 403 was classified as a dead grant. classifyProviderError mapped
any 401 or 403 to PROVIDER_AUTH_EXPIRED, which is fatal, so a Fortnox account
without leverantorsregister permission aborted the whole migration at the
suppliers step with "Anslutningen har gatt ut. Ateranslut" even though the same
token had just succeeded on the previous step. Reconnecting can never fix that,
and steps 4 and later never ran. The provider's own reason ("Saknar behorighet
for leverantorsregister.") never reached the user. A 403 is now non-fatal once
the same token has already succeeded in the run, the migration continues, and
the provider's reason is surfaced. A 401, or a 403 on the first call, keeps the
auth-expired path.
fetchCompanyInfoDirect swallowed every error and returned null, which made the
existing PROVIDER_API_MODULE_INACTIVE remediation unreachable: a Visma customer
whose api_standard module is off got a silent 200 with an empty company card
instead of the precise Swedish explanation that was already written.
Kreditfakturor were dropped entirely. entity-mapper wrote document_type
'credit_note', but invoices_document_type_check allows only invoice, proforma
and delivery_note, and credit notes are modelled by credited_invoice_id. Every
migrated kreditfaktura was rejected and counted as skipped. One customer
imported 255 sales invoices and 0 credit notes on 2026-08-31; AR and revenue
are overstated by the credited amounts, and kreditfakturor are
rakenskapsinformation. They now import as invoice rows with reversed amounts
and status 'credited', following the in-app credit convention. They import
unlinked: no provider DTO carries a reference to the invoice being credited, so
there is nothing to match on and guessing would corrupt the AR ledger. The
wizard says so instead of burying them in skipped.
Also makes the OAuth callback non-replayable from browser history (no-store
plus history replacement), which is what the "state rejected" events were: a
replay of a callback that had already succeeded seconds earlier. No
already-connected page, so consumed-vs-unknown state stays unobservable to an
unauthenticated caller. Expected PSD2 session expiry drops from error to warn.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016ifKg6Ec67A39oxfGPU1yc
* fix(arcim): entity line needs the failed flag
The unlinked-credit-note row omitted `failed`, which the entityLines element
type requires. Caught by the zero-extensions build, not by vitest: the unit
suite does not typecheck.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016ifKg6Ec67A39oxfGPU1yc
* fix(arcim): write the missing-reference disclosure onto the credit note itself
Review finding (swedish-compliance-review-bot): ML 17 kap 22-23 § wants a
kreditfaktura to reference the invoice it credits, and BFL 5 kap 6-7 § wants a
verifikation to reference its underlag. No provider DTO carries that reference,
so the pairing cannot be resolved at import and guessing it would corrupt the
AR ledger. Reporting the count in the migration wizard is not enough: a result
screen is not rakenskapsinformation, and the gap has to be legible on the
record itself years later.
The disclosure now goes into invoices.notes and supplier_invoices.notes,
preserving whatever note the provider sent.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016ifKg6Ec67A39oxfGPU1yc
---------
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
Jakob Wennberg
|
088754d61a |
fix(processing-history): register the missing event types, and strip the PII two of them carry (#2111)
* fix(processing-history): register the missing event types, and strip the PII two of them carry Ten event types are emitted by code but absent from processing_event_types, so every append fails the foreign key. Appends are best-effort try/catch, so no user request fails, but the internal audit trail is empty for ten kinds of legally motivated act, including the BFL 5 kap 5 § rattelse record when a user swaps a transaction's underlag (TransactionDocumentReplaced) and the SOC 2 revocation record (OAuthClientRevoked). Order matters and is deliberate. Two invoice-inbox events, RateLimitedDropped and AttachmentsTruncated, put the raw sender address and mail subject in their payload. Registering those types first would start persisting that PII into an append-only table whose UPDATE is trigger-blocked and which the archive's erasure path excludes. The strip therefore ships in this same commit, ahead of the migration. Only the invoice-inbox emitter was edited. whatsapp-inbox shares the RateLimitedDropped type name with a payload that carries no phone number. Closes the class rather than the two logged instances: a TypeScript union makes an unregistered literal a compile error, and the pg test asserts the database catalog is a superset of the code's list, generated from the union. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016ifKg6Ec67A39oxfGPU1yc * fix(processing-history): strip the inbound-mail PII in the database, not by deploy ordering Review finding (superagent-security, P2): shipping the emitter fix and the catalog migration in one commit is not the same as one instant. Migrations apply on merge while the replacement build takes minutes, so an old instance can still write a sender address and mail subject in that window, and such a row is permanent: processing_history takes no UPDATE and no DELETE, and the archive export excludes it from the erasure path. Adds a BEFORE INSERT trigger stripping `from` and `subject` from the RateLimitedDropped and AttachmentsTruncated payloads, and keeps it afterwards so the invariant belongs to the table rather than to one emitter's good behaviour. The jsonb object check is load bearing: `payload - 'key'` raises on a jsonb array and payload's shape is not constrained. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016ifKg6Ec67A39oxfGPU1yc --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Mattsson
|
08b1119c7d |
feat(connect): wire the SKV extension through the connector broker + data proxy (PR6b-2) (#2103)
* feat(connect): wire the SKV extension through the connector broker + data proxy (PR6b-2) In connector mode (GNUBOK_CONNECTOR_KEY set, no own SKV credentials) the Skatteverket extension now routes through the hosted connector stack (#1757) instead of calling Skatteverket directly: - skvRequestWithAuth routes to the data proxy: base URL maps to a service segment (moms/skattekonto/agd-inlamning/agd-period), the user's SKV Bearer moves to X-Connector-Upstream-Authorization, the connector key authenticates the proxy, and the gateway Client_Id/Client_Secret are omitted (the proxy adds Arcim's). Connector-layer 4xx bodies (code CONNECTOR_*) are classified before the SKV-shaped 401/403 sniffing so a broker refusal surfaces operator guidance (check GNUBOK_CONNECTOR_KEY), never APIGW/BankID guidance for knobs the instance does not have. - OAuth: /authorize starts the consent via the broker's authorize-url (persisting its redirect_uri + connector_state), the hosted SKV callback bounces the code back to the instance, and exchangeCodeForTokens / refreshAccessToken exchange through the broker's /oauth/token, unwrapping its { data } envelope. Tokens still rest encrypted on the instance; client_id/client_secret never exist there. - Broker refresh 404 CONNECTOR_NOT_OWNED maps to SESSION_EXPIRED (terminal; reconnect fixes); broker 502 stays a raw error so a transient SKV outage never re-arms the reconnect banner (#1155). - getSkatteverketEnvironment() reports 'prod' in connector mode: the upstream env is hosted's, and the instance's unset defaults would show a false Testmiljo badge on real filings. - System (CCG/ombud) auth is deliberately not brokered: hosted-only, stays direct. Hosted and own-credentials self-hosts are byte-identical: every branch gates on skatteverketConnectorMode(), which is null whenever own SKV credentials exist or no connector key is set. Direct-path tests pin that. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KRfamAKDqvRNwbjr5XD2VS * fix(connect): classify SKV dead-refresh-token dialects broker-side; forward diagnostic headers; connector-aware gateway guidance Skeptic refutation on PR #2103 (found independently by the correctness and compliance skeptics): the broker's /oauth/token catch-all collapsed SKV's terminal dead-refresh-token dialects (404 id_not_found, 400 invalid_grant, "Refresh Token status is expired": the dominant refresh outcome, per-flow tokens live 65 minutes) into the generic 502 CONNECTOR_SKV_TOKEN_FAILED, so a connector instance could never classify ordinary session expiry: raw English 500s instead of the reconnect flow, staged filing operations consumed as non-recoverable, crons retrying raw forever. - Broker /oauth/token: re-codes those dialects as 401 CONNECTOR_SKV_REFRESH_DEAD, refresh grant only (invalid_grant on the code exchange means an expired one-shot code and keeps the generic 502). The classifier (isSkvDeadRefreshTokenError) uses the same regex set the extension's direct path classifies with. - Instance dead-token classifier maps CONNECTOR_SKV_REFRESH_DEAD to SESSION_EXPIRED alongside 404 CONNECTOR_NOT_OWNED; the generic 502 stays a raw error so a transient SKV outage never re-arms the reconnect banner. - Data proxy: forwards WWW-Authenticate and x-skv-*/x-amzn-*/x-api-* response headers (the instance's MISSING_SCOPE classification reads them; body-less gateway rejections carry no other signal). - Instance gateway-refusal guidance is connector-aware: a self-host has no SKATTEVERKET_APIGW_CLIENT_ID and no Utvecklarportalen access, so connector mode points at /api/connector/status and support instead. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KRfamAKDqvRNwbjr5XD2VS * fix(connect): reject redirects on the instance's broker OAuth requests CodeRabbit inline finding (CWE-200): the connector-mode authorize-url and token requests followed redirects by default, so a 307/308 would resend the connector key (and code/refresh token) to the redirect target. redirect 'error', matching the broker's own postToken rule; the token response must only ever come from the broker endpoint itself. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KRfamAKDqvRNwbjr5XD2VS --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
aabddb592f |
feat(billing): multi-user paywall: multi_user capability, 20-day grace, owner-only dormancy (#2099)
* feat(billing): multi-user seat gate: multi_user capability, 20-day grace, owner-only dormancy Multiple people in one company becomes a paid capability (multi_user, the eighth PAID key). Derived at access time from capability_grants, no status column, no enforcement cron: - entitled: active grant (trial/stripe/team/manual/comp), everyone works - grace: newest grant expired < 20 days ago; countdown banner for everyone in companies with > 1 user; invites still allowed - frozen: only role=owner resolves; other memberships go dormant (rows untouched, paying reactivates instantly); invites 403 with paid-plan upsell Enforcement: new resolve_active_company_gated RPC (zero-arg RPC and RLS twin untouched: they also run on self-hosts, where the gate never bites), gated query fallback for service-role/API-key paths, setActiveCompany guard, MCP company-access check, invite route. Middleware routes all-frozen users to a new /paused page; the switcher greys locked companies. Migration 20260901081417 (applied to staging): trial trigger seeds multi_user, backfills for mid-trial companies, active Stripe subs, team agreements, and a grandfather grant (expires now, i.e. grace = deploy + 20 days) for existing unpaid multi-member companies. Daily cron mails owners at grace start and last day. Strings in sv+en; pg-real + unit tests included. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01L4tNt8wRG3a5iuU1JE2pnP * fix(billing): multi-user seat gate hardening from skeptic review - Stripe cancel now EXPIRES the multi_user stripe grant instead of deleting it: the 20-day grace window hangs on an expired row, so a deleted one froze churned payers' staff instantly with no banner and no mail. Other stripe grants keep the freeze-and-retain delete. - New SECURITY DEFINER company_multi_user_state() RPC (migration 20260901083726, applied to staging) and RPC-first getMultiUserState: capability_grants RLS hides team-scoped rows from non-team users, so user-client reads misread byra-covered companies as frozen (switch refusal, wrong switcher locks). - Byra-kind teams get a standing team-scoped multi_user grant (backfill + teams trigger): byra client companies have no company-scoped trial by design, so a grantless byra team would freeze every consultant and client user. - Comped/manual companies with active PAID-key grants extend to multi_user (a comped company must not read as paying while locking out user two). - /api/v1 gets the same dormancy gate as MCP (frozen non-owner -> 403). - PGRST202 on resolution fails OPEN (pre-migration DB has zero multi_user rows; the gated fallback would have frozen every non-owner mid-deploy). - Grace cron: covers team-scoped lapses (byra agreement ending) and skips the start mail for the hand-mailed grandfather cohort. - Tests updated/added across all touched surfaces; pg tests for the new RPC and byra trigger; trial-suppression pg test extended to 8 keys. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01L4tNt8wRG3a5iuU1JE2pnP * fix(billing): decouple seat-gate env check and fail open on gate read throws CI round 1 on #2099: - isMultiUserEnforced no longer imports has-capability: several route test suites partially mock that module and the vitest mock guard threw from inside the v1 seat gate, turning expected 4xx responses into 500s. multi_user is never a connector capability, so the bypass reduces to the same env reads, now inlined. - getMultiUserState wraps its resolution in a fail-open try/catch: a client without .rpc or a thrown network error must never lock users out. - no-phantom-columns ceiling 391 -> 393 with reasons: the seat gate's .or() scope filter (server-resolved UUIDs) and the Stripe cancel expiry update's timestamp .or(); all columns in both strings are literals. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01L4tNt8wRG3a5iuU1JE2pnP * fix(billing): membership-guard the multi-user entitlement RPCs (Superagent P3) company_multi_user_ok and company_multi_user_state are SECURITY DEFINER and were granted to authenticated with a caller-supplied company UUID: any logged-in user could probe an arbitrary company's billing state and grace deadline across tenants. Migration 20260901091752 (applied to staging) requires an auth.uid() membership in the target company when a JWT is present, keeps service-role/definer contexts unrestricted, and clamps the grace window to [0, 20] days. pg tests: stranger gets false/NULL, member reads normally, oversized p_grace_days cannot widen the probe. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01L4tNt8wRG3a5iuU1JE2pnP --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Mattsson
|
05dce83a2b |
feat(connect): Enable Banking client routes through the hosted proxy in connector mode (PR6b-1) (#2094)
* feat(connect): route the Enable Banking client through the hosted proxy in connector mode PR6b-1 of the instance-side client wiring. Until now bankConnectorMode() had no consumer but the status label; this makes a self-host with a connector key and no own EB credentials actually reach Enable Banking through the hosted bank proxy. - api-client authenticatedFetch: in connector mode swap the base URL to the proxy and send the connector key as a Bearer token. The EB JWT signer (getAuthorizationHeader) is never called: the instance holds no private key. On hosted and on own-credentials self-hosts the direct path is byte-identical. - startAuthorization forwards X-Connector-Company so the proxy can meter the per-company connection quota; index.ts passes companyId at both connect sites. - createSession forwards the signed connector_state so the proxy binds the /sessions exchange to the pending ledger row (single-use, race-safe). - callback route reads connector_state from the query (echoed by the hosted callback) and threads it through finalizeConnection. Tests: connector-mode base/auth/company-header/connector_state assertions in api-client, direct-path and own-credentials byte-identity, and the callback threading both connector and direct paths. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): gate X-Connector-Company on connector mode, not on companyId Skeptic regression finding: index.ts passes companyId to startAuthorization unconditionally, and the header was attached whenever companyId was truthy. On hosted and on own-credentials self-hosts companyId is always set, so every direct POST /auth to the real Enable Banking API carried the tenant's internal company UUID: a needless behavior change on the production path and an identifier leak to a third-party PSD2 processor (the "byte-identical direct path" claim was false). Gate the header on bankConnectorMode() so it is sent only when the request actually goes to the hosted proxy. Adds a direct-path test asserting the header is absent even when companyId is passed. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
b643e6ce64 |
perf(mcp): demote ten unused catalog reads, reclaiming 3 763 tokens of budget (#2089)
* perf(mcp): demote ten unused catalog reads, reclaiming 3 763 tokens of budget The tools/list payload bench sat about 20 tokens under its 65 000 ceiling, so the next tool or field anyone added failed CI. That is not hypothetical: it already cost the agent-briefing field in #2079, which was built and then removed for want of roughly 85 tokens. The bench's own comments deferred the fix twice, in the same words both times: picking which read to demote needs prod usage data, not a guess inside an unrelated PR, so do the demotion as its own change and ratchet the ceiling back down. This is that change. Selection is from 60 days of mcp.tool_called: default-catalog READ tools with 25 or fewer calls, excluding anything named in RECOMMENDED_WORKFLOW_LOADOUTS. Ten qualified. Measured 65 046 -> 61 283, ceiling ratcheted to 61 600. Usage data is necessary but not sufficient, and three classes were kept in the catalog despite low counts: - gnubok_call_tool, which IS the search-only bridge. - the connect_* onboarding tools and gnubok_lookup_company. A fresh agent has not learned to search yet, which is exactly why #1936 put them in the catalog. - every arsredovisning, dispositioner, depreciation and accrual PROPOSAL tool. A 60-day window ending in August cannot see bokslut season: most Swedish companies close on 31 December and do the work January to June, so summer counts understate these to near zero. Demoting them would hide the year-end flow precisely when it is needed. Widget tools are a hard exclusion, and I found that by tripping over it: gnubok_receipt_matcher and gnubok_vat_review_widget were demoted in a first pass and their own suites failed. A widget is rendered from the _meta.ui it publishes in tools/list, so search-only leaves it callable but unrenderable. Both are back in the catalog, and a general invariant now holds the rule for every widget tool added later, verified by demoting one and watching it fail. Only READ tools are demotable at all: gnubok_call_tool refuses writes, so a search-only WRITE is uncallable on Claude.ai. Demoted reads stay callable through the bridge; they are no longer listed up front. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs: code-span the connect_* identifier in DECISIONS A bare asterisk opens an emphasis span in Markdown. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
37e50c272d |
feat(connect): Skatteverket broker + data proxy for self-hosted instances (tokens stay on the instance) (#1757)
* feat(entitlements): partition the self-host bypass so connector capabilities fall through to grants; capability_grants.source accepts 'connector' Sovereign plan WS3 PR3: ships dark, nothing changes for hosted. - lib/entitlements/keys.ts: CONNECTOR_CAPABILITIES = bank_sync, skatteverket, org_lookup, migration (services Accounted operates that a self-hosted instance cannot provide itself) + isConnectorCapability(). Separate from PAID_CAPABILITIES and outside the trial-seed trigger on purpose: a hosted company can never hold a connector grant. - lib/entitlements/has-capability.ts: isPaywallBypassed() -> isBypassedFor(key). Hosted: byte-identical (dev / DISABLE_PAYWALL bypass, FORCE_PAYWALL wins, else the grant lookup). Self-host: local capabilities always on (FORCE_PAYWALL included, as the existing test demands); connector capabilities behave like hosted, i.e. dev bypass, FORCE_PAYWALL, else the grant lookup where the connector sync will write source='connector' rows. getCompanyEntitlements on a self-host: local paid keys + active connector keys, state 'paid' with an active connector grant else 'none' (never the hosted trial copy). - Migration 20260820122000: capability_grants.source CHECK gains 'connector', found through pg_constraint (the CHECK was declared inline and auto-named; Postgres stores IN as = ANY, matched accordingly). pg-real test: connector accepted, unknown source rejected, upsert on the (scope, key, source) identity, trial seed writes no connector rows. - Tests: self-hosted connector matrix (local all-on without DB, connector gated by grant/expiry, dev bypass all-on, FORCE_PAYWALL gates connector keys only, bulk resolution, entitlements shape); two pre-existing tests that asserted the old "self-host holds connector keys" contract updated to the new one. Verified: full unit suite green, pg-real suite for lib/entitlements green against a local supabase/postgres with every migration applied, lint ratchet, guards. Deferred to the instance-wiring PR: adding the connector extensions to the self-host Docker preset (dead-end upsells until a key can be issued). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(entitlements): fold the self-host branch into the existing grants query One .or(scopeFilter), not two: the duplicated helper pushed the no-phantom-columns unresolvable-expression count to 380/379. Behaviour is unchanged; the self-host matrix tests still pass. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(connect): hosted connector-key registry + validate RPC + entitlements endpoint; instance sync writes connector grants hourly Sovereign plan WS3 PR4 ("key infra enabling manual sales"), stacked on the entitlement partition (#1747). Nothing is purchasable yet; this is the plumbing both ends need before the first manually issued key. Hosted side: - Migration 20260820123000: connector_keys (SHA-256 key_hash, prefix, org_number, pinned instance_url, scopes, status, Stripe ids, current_period_end, per-minute rate limit, active_company_count, last_seen/synced) and connector_usage_events (per-request metering, separate from metered_events whose company_id references hosted companies). RLS on, NO policies: service role only. RPC validate_and_increment_connector_key copies the api_keys pattern (FOR UPDATE, minute window, suspended reported not counted, revoked = no row) and is REVOKEd from PUBLIC/anon/authenticated, GRANTed to service_role. pg-real test covers validate/count, unknown+revoked, suspended, rate limit, execute privileges per role, RLS invisibility, usage cascade. - lib/connect/contract.ts (shared wire types), lib/connect/hosted/keys.ts (generate/hash/validate -> 401/403/429 mapping), with-connector-auth.ts (Bearer or X-Connector-Key, one usage row per request, 500 envelope on handler throw), /api/connect/entitlements GET + POST (records active_company_count, pins instance_url on first report, never moves a pinned one), scripts/issue-connector-key.ts (dry run unless --confirm, prints the key once + the .env lines). Instance side: - lib/connect/instance/config.ts (GNUBOK_CONNECTOR_KEY, GNUBOK_CONNECT_URL default https://app.gnubok.se), sync.ts: reports the active company count and writes source='connector' grants for every company x covered scope, expires_at = min(now+72h, period_end+3d); 401/403 or a non-active status deletes them (freeze-and-retain); network/5xx/429 leave them alone. /api/connector/sync/cron (hourly) runs it; not_configured without a key. - Crontab generator gains EXTRA_JOBS (variant-only jobs not in vercel.json, with reasons) + drift tests; docker/crontab.self-hosted regenerated with the hourly sync. Docs (SELF-HOSTING connector section, env templates), DECISIONS. Tests: 52 new unit tests (keys, auth wrapper, route, config, sync outcomes and grant arithmetic, cron route, crontab EXTRA_JOBS) + 7 pg-real tests run locally against supabase/postgres with every migration applied. no-phantom-columns ceiling +1 with a reason (the bulk grant upsert). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(connect): Enable Banking proxy for self-hosted instances, with a secret-free ownership ledger and a global rate budget Sovereign plan WS3 PR5a, stacked on the connector-key infra (#1748). A self-hosted instance with a `bank_sync`-scoped connector key can now connect a bank through Arcim's PSD2 credentials; the bank session id and all transaction data stay in the instance's own database (founder decision: tokens on the instance, proxy stateless). - Migration 20260820124000: `connector_connections` (secret-free ledger: sha256 of the EB session id + account uids, service-role only), `connector_upstream_counters` + RPC `connector_reserve_upstream` (global budget under EB Annex 1 §5's 300/min, shared with hosted), and `connector_keys.limits` jsonb; validate RPC v2 returns limits. All RPCs REVOKEd from PUBLIC/anon/authenticated, GRANTed service_role. pg-real covers all of it. - EB JWT minting moved to lib/connect/upstreams/enable-banking-jwt.ts (core must not import @/extensions/); the extension re-exports it, tests unchanged. - lib/connect/hosted/{state,ledger,upstream-budget}.ts: HMAC-signed connector state (15-min TTL) so the consent redirect can use OUR registered EB callback and bounce back to the instance, no per-instance redirect URI at EB; the callback route gains that connector branch. - app/api/connect/bank/[...path]: path allowlist (aspsps, auth, sessions, accounts/{uid}/{balances,transactions}), never open passthrough. POST /auth enforces the per-company connection quota + rewrites redirect/state; reads/deletes verify ledger ownership; every upstream call takes the global budget (429 + Retry-After when exhausted). - issue-connector-key.ts: scopes default bank_sync,skatteverket (TIC out of v1), --bank/skv-connections-per-company + --sync-min-interval. - Docs (SELF-HOSTING: bank connector live), DECISIONS. Verified: 52 connect unit tests + 13 pg-real (run locally against supabase/postgres with all migrations) + EB extension suite (225, jwt relocation intact); full unit suite 15 979 green; tsc, guards, lint clean. Not in this PR: SKV broker (PR5b) and instance wiring (PR6). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(connect): Skatteverket broker + data proxy for self-hosted instances (tokens stay on the instance) Sovereign plan WS3 PR5b, stacked on the bank proxy (#1751). A self-hosted instance with a `skatteverket`-scoped connector key can now run the BankID consent, file VAT/AGI and sync skattekonto through Arcim's registered Skatteverket client; the SKV tokens are returned to the instance and stored (encrypted) there. - lib/connect/upstreams/skatteverket-oauth.ts: core-side SKV OAuth + data helpers (authorize URL, code/refresh exchange with Arcim's client secret, the four backing-API base URLs, the API-gateway Client_Id/Client_Secret headers). Core can't import @/extensions/, so this duplicates the extension's endpoints/scope set (one integrator = Arcim), mirroring the EB JWT relocation. - app/api/connect/skv/oauth/authorize-url: builds the authorize URL against OUR registered redirect_uri + a signed connector state, per-company SKV connection quota, pending ledger row. - app/api/connect/skv/oauth/token: exchanges/refreshes and RETURNS the tokens to the instance; the ledger keeps only sha256(access_token) + sha256(refresh_token). - app/api/connect/skv/api/[...path]: allowlist over moms / skattekonto / agd-inlamning / agd-period. The instance sends the user's SKV Bearer (as X-Connector-Upstream-Authorization) + X-Connector-Key; the proxy checks the token hash against the ledger, adds Arcim's gateway credentials (never exposed to the instance), forwards. Same per-key + global budget as bank. - The Skatteverket extension /callback gains the connector branch (isConnectorState -> 302 back to the instance; code never exchanged there). - Docs (SELF-HOSTING: SKV connector live) + DECISIONS. Tests: SKV oauth lib, authorize-url, token, data proxy, callback connector branch (all green; 74 connect + 425 connect/SKV). tsc, guards, lint clean; no-phantom-columns held at 380 (literal update branches). Not in this PR: instance-side wiring (PR6). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(connect): SKV broker review+skeptic batch: state-bound exchange, owned-only refresh, identity-number redaction, quota reservation, https-only bases, docs dedupe Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * docs(self-host): restore the instance-wiring qualifier in the connector section Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): PR #1757 review batch 2: redirect 'error' on credential fetches, mandatory ledger writes before token return, MD037 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): reject encoded path separators in SKV data-proxy segments (traversal guard) The WHATWG parser normalizes raw dot segments before the route runs; what survives is an encoded separator inside a segment (a%2Fb, ..%2Fx, a%5Cb), which would escape the allowlisted service base once the upstream fetch re-normalizes. splitPath now decodes each segment and rejects dot segments and separator-bearing values. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: Emil <emilmattsson14@gmail.com> |
||
|
Jakob Wennberg
|
10bbfb9d79 |
fix(mcp): make agent failures diagnosable from the log, not just from the source (#2087)
* fix(mcp): make agent failures diagnosable from the log, not just from the source Mining event_log for mcp.tool_called: VALIDATION_ERROR ran at about one a day until 2026-08-25, then jumped to a hundred a day and stayed there. One integration's gnubok_get_kpi_report has been refused 604 times over seven days and is still failing. The cause is the unknown-parameter guard from #1856, which is correct and stays. The caller is even told precisely what is wrong: getStructuredError puts "Unknown parameter "x" for <tool>. Valid parameters: ..." into message_en, which the agent receives. What was broken is what we recorded about it. Two things, both cheap: - Telemetry logged only message_sv, which for VALIDATION_ERROR is the registry default "Förfrågan innehåller ogiltiga uppgifter." and names neither the parameter nor the tool. A seven-day outage was indistinguishable from a typo, and the cause was only findable by reading the dispatcher source. errorDetail now carries message_en, stored only when it differs from errorMessage, so the many domain failures whose message_sv is already the specific text cost nothing extra. - errorKind said 'company_access_denied' for all 604, because the arg guard throws inside the company-routing try. That is an active misdirection: it sends triage looking for a tenancy bug that does not exist. Exactly two things in that block raise VALIDATION_ERROR, the unknown-parameter guard and a malformed company_id, and neither is a permissions failure, so they now log as 'invalid_arguments'. Tests reproduce the production call shape and were verified to fail when either fix is reverted. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(mcp): derive the telemetry payload type from the event contract Two review findings, both valid. The local ToolCalledPayload interface was a hand-maintained duplicate that omitted sessionId and half the errorKind union. That is not cosmetic: it is why errorDetail could be added to the emitter and to lib/events/types.ts while the test file type-checked against a stale shape. Deriving it from EventPayload<'mcp.tool_called'> removes the drift. The null-detail assertion was also conditional, so it would have passed on a wrong-but-present value. Replaced with two determinate cases: a scope denial carries both languages without duplicating either, and the unknown-tool exit, which supplies no diagnostic, stores null. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(mcp): pin the scope name in the diagnostic assertion 'A different string' passes on any placeholder. The reason errorDetail is worth storing is that it names what the caller lacks, so assert that. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
523beaa18b |
fix(mcp): give the missing-chart-accounts refusal a code agents can dispatch on (#2075)
getStructuredError resolves an error with no `code` to UNKNOWN_ERROR, whose registry text is the constant "Något gick fel. Försök igen." So an agent branching on `code` saw "unknown" for a failure whose own message already named the accounts and the two tools that fix it. On production that was 40 of the create_voucher failures in 60 days. Everything needed already existed: the ACCOUNTS_NOT_IN_CHART code, its registry entry, its remediation pointing at the chart-of-accounts resource, and a typed error class that storno-service already throws. This path just never attached the code. Attached with Object.assign rather than by throwing AccountsNotInChartError, because that class builds its own generic English message from the account list and would discard the richer Swedish one, which is the more useful half. Pinned by a test asserting the same message WITHOUT the code still resolves to UNKNOWN_ERROR, so this cannot silently regress. Not fixed here: the MCP server has ~155 bare Swedish-prose throws against 3 uses of codedError(), so most domain failures remain undispatchable. This fixes the one instance telemetry actually proved rather than converting 155 sites blind. Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Jakob Wennberg
|
36123cef23 |
feat(connect): Enable Banking proxy for self-hosted instances, with a secret-free ownership ledger and a global rate budget (#1751)
* feat(entitlements): partition the self-host bypass so connector capabilities fall through to grants; capability_grants.source accepts 'connector' Sovereign plan WS3 PR3: ships dark, nothing changes for hosted. - lib/entitlements/keys.ts: CONNECTOR_CAPABILITIES = bank_sync, skatteverket, org_lookup, migration (services Accounted operates that a self-hosted instance cannot provide itself) + isConnectorCapability(). Separate from PAID_CAPABILITIES and outside the trial-seed trigger on purpose: a hosted company can never hold a connector grant. - lib/entitlements/has-capability.ts: isPaywallBypassed() -> isBypassedFor(key). Hosted: byte-identical (dev / DISABLE_PAYWALL bypass, FORCE_PAYWALL wins, else the grant lookup). Self-host: local capabilities always on (FORCE_PAYWALL included, as the existing test demands); connector capabilities behave like hosted, i.e. dev bypass, FORCE_PAYWALL, else the grant lookup where the connector sync will write source='connector' rows. getCompanyEntitlements on a self-host: local paid keys + active connector keys, state 'paid' with an active connector grant else 'none' (never the hosted trial copy). - Migration 20260820122000: capability_grants.source CHECK gains 'connector', found through pg_constraint (the CHECK was declared inline and auto-named; Postgres stores IN as = ANY, matched accordingly). pg-real test: connector accepted, unknown source rejected, upsert on the (scope, key, source) identity, trial seed writes no connector rows. - Tests: self-hosted connector matrix (local all-on without DB, connector gated by grant/expiry, dev bypass all-on, FORCE_PAYWALL gates connector keys only, bulk resolution, entitlements shape); two pre-existing tests that asserted the old "self-host holds connector keys" contract updated to the new one. Verified: full unit suite green, pg-real suite for lib/entitlements green against a local supabase/postgres with every migration applied, lint ratchet, guards. Deferred to the instance-wiring PR: adding the connector extensions to the self-host Docker preset (dead-end upsells until a key can be issued). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(entitlements): fold the self-host branch into the existing grants query One .or(scopeFilter), not two: the duplicated helper pushed the no-phantom-columns unresolvable-expression count to 380/379. Behaviour is unchanged; the self-host matrix tests still pass. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(connect): hosted connector-key registry + validate RPC + entitlements endpoint; instance sync writes connector grants hourly Sovereign plan WS3 PR4 ("key infra enabling manual sales"), stacked on the entitlement partition (#1747). Nothing is purchasable yet; this is the plumbing both ends need before the first manually issued key. Hosted side: - Migration 20260820123000: connector_keys (SHA-256 key_hash, prefix, org_number, pinned instance_url, scopes, status, Stripe ids, current_period_end, per-minute rate limit, active_company_count, last_seen/synced) and connector_usage_events (per-request metering, separate from metered_events whose company_id references hosted companies). RLS on, NO policies: service role only. RPC validate_and_increment_connector_key copies the api_keys pattern (FOR UPDATE, minute window, suspended reported not counted, revoked = no row) and is REVOKEd from PUBLIC/anon/authenticated, GRANTed to service_role. pg-real test covers validate/count, unknown+revoked, suspended, rate limit, execute privileges per role, RLS invisibility, usage cascade. - lib/connect/contract.ts (shared wire types), lib/connect/hosted/keys.ts (generate/hash/validate -> 401/403/429 mapping), with-connector-auth.ts (Bearer or X-Connector-Key, one usage row per request, 500 envelope on handler throw), /api/connect/entitlements GET + POST (records active_company_count, pins instance_url on first report, never moves a pinned one), scripts/issue-connector-key.ts (dry run unless --confirm, prints the key once + the .env lines). Instance side: - lib/connect/instance/config.ts (GNUBOK_CONNECTOR_KEY, GNUBOK_CONNECT_URL default https://app.gnubok.se), sync.ts: reports the active company count and writes source='connector' grants for every company x covered scope, expires_at = min(now+72h, period_end+3d); 401/403 or a non-active status deletes them (freeze-and-retain); network/5xx/429 leave them alone. /api/connector/sync/cron (hourly) runs it; not_configured without a key. - Crontab generator gains EXTRA_JOBS (variant-only jobs not in vercel.json, with reasons) + drift tests; docker/crontab.self-hosted regenerated with the hourly sync. Docs (SELF-HOSTING connector section, env templates), DECISIONS. Tests: 52 new unit tests (keys, auth wrapper, route, config, sync outcomes and grant arithmetic, cron route, crontab EXTRA_JOBS) + 7 pg-real tests run locally against supabase/postgres with every migration applied. no-phantom-columns ceiling +1 with a reason (the bulk grant upsert). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(connect): Enable Banking proxy for self-hosted instances, with a secret-free ownership ledger and a global rate budget Sovereign plan WS3 PR5a, stacked on the connector-key infra (#1748). A self-hosted instance with a `bank_sync`-scoped connector key can now connect a bank through Arcim's PSD2 credentials; the bank session id and all transaction data stay in the instance's own database (founder decision: tokens on the instance, proxy stateless). - Migration 20260820124000: `connector_connections` (secret-free ledger: sha256 of the EB session id + account uids, service-role only), `connector_upstream_counters` + RPC `connector_reserve_upstream` (global budget under EB Annex 1 §5's 300/min, shared with hosted), and `connector_keys.limits` jsonb; validate RPC v2 returns limits. All RPCs REVOKEd from PUBLIC/anon/authenticated, GRANTed service_role. pg-real covers all of it. - EB JWT minting moved to lib/connect/upstreams/enable-banking-jwt.ts (core must not import @/extensions/); the extension re-exports it, tests unchanged. - lib/connect/hosted/{state,ledger,upstream-budget}.ts: HMAC-signed connector state (15-min TTL) so the consent redirect can use OUR registered EB callback and bounce back to the instance, no per-instance redirect URI at EB; the callback route gains that connector branch. - app/api/connect/bank/[...path]: path allowlist (aspsps, auth, sessions, accounts/{uid}/{balances,transactions}), never open passthrough. POST /auth enforces the per-company connection quota + rewrites redirect/state; reads/deletes verify ledger ownership; every upstream call takes the global budget (429 + Retry-After when exhausted). - issue-connector-key.ts: scopes default bank_sync,skatteverket (TIC out of v1), --bank/skv-connections-per-company + --sync-min-interval. - Docs (SELF-HOSTING: bank connector live), DECISIONS. Verified: 52 connect unit tests + 13 pg-real (run locally against supabase/postgres with all migrations) + EB extension suite (225, jwt relocation intact); full unit suite 15 979 green; tsc, guards, lint clean. Not in this PR: SKV broker (PR5b) and instance wiring (PR6). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore(connect): update ledger pg test to re-versioned migration 20260831200000 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): redact opaque path segments before usage metering; correct stale RPC-source comment GET/DELETE /sessions/{id} and /accounts/{uid}/... carry the raw EB session id / account uid in the pathname; metering persisted it in cleartext next to the ledger that stores only sha256(handle). Opaque segments (UUID, long hex, long base64url) now become ':id' before the connector_usage_events insert. Migration comment now cites the real prior RPC source (20260831190000). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): percent-encoded path segments count as opaque in metering redaction Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): PR #1751 review batch: https-only EB URL, body-covering timeout, quota reservation, delete-after-success, doc fix Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(connect): bind the /sessions code exchange to its verified pending state; ceiling +1 Verified state signature, key/service match, and an existing pending row now precede the EB exchange; a concurrently consumed state closes the just-minted upstream session and 409s. no-phantom-columns ceiling 391 for countHeldConnections' computed .or() timestamp filter. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: Emil <emilmattsson14@gmail.com> |
||
|
Jakob Wennberg
|
cfce2de925 |
feat(entitlements): partition the self-host bypass so connector capabilities fall through to grants; capability_grants.source accepts 'connector' (#1747)
* feat(entitlements): partition the self-host bypass so connector capabilities fall through to grants; capability_grants.source accepts 'connector' Sovereign plan WS3 PR3: ships dark, nothing changes for hosted. - lib/entitlements/keys.ts: CONNECTOR_CAPABILITIES = bank_sync, skatteverket, org_lookup, migration (services Accounted operates that a self-hosted instance cannot provide itself) + isConnectorCapability(). Separate from PAID_CAPABILITIES and outside the trial-seed trigger on purpose: a hosted company can never hold a connector grant. - lib/entitlements/has-capability.ts: isPaywallBypassed() -> isBypassedFor(key). Hosted: byte-identical (dev / DISABLE_PAYWALL bypass, FORCE_PAYWALL wins, else the grant lookup). Self-host: local capabilities always on (FORCE_PAYWALL included, as the existing test demands); connector capabilities behave like hosted, i.e. dev bypass, FORCE_PAYWALL, else the grant lookup where the connector sync will write source='connector' rows. getCompanyEntitlements on a self-host: local paid keys + active connector keys, state 'paid' with an active connector grant else 'none' (never the hosted trial copy). - Migration 20260820122000: capability_grants.source CHECK gains 'connector', found through pg_constraint (the CHECK was declared inline and auto-named; Postgres stores IN as = ANY, matched accordingly). pg-real test: connector accepted, unknown source rejected, upsert on the (scope, key, source) identity, trial seed writes no connector rows. - Tests: self-hosted connector matrix (local all-on without DB, connector gated by grant/expiry, dev bypass all-on, FORCE_PAYWALL gates connector keys only, bulk resolution, entitlements shape); two pre-existing tests that asserted the old "self-host holds connector keys" contract updated to the new one. Verified: full unit suite green, pg-real suite for lib/entitlements green against a local supabase/postgres with every migration applied, lint ratchet, guards. Deferred to the instance-wiring PR: adding the connector extensions to the self-host Docker preset (dead-end upsells until a key can be issued). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * refactor(entitlements): fold the self-host branch into the existing grants query One .or(scopeFilter), not two: the duplicated helper pushed the no-phantom-columns unresolvable-expression count to 380/379. Behaviour is unchanged; the self-host matrix tests still pass. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(entitlements): self-host connector gate honors only source=connector grants The trial-seed trigger (seed_trial_capability_grants) writes 30-day source='trial' rows for bank_sync and skatteverket on every company insert, self-hosts included. The partitioned self-host gate read every active grant, so a fresh self-host company held every connector capability for a month with no connector key (CodeRabbit finding on #1747, verified against migration 20260818170000). hasCapability and getCompanyIdsWithCapability now add .eq('source', 'connector') on a self-host; getCompanyEntitlements skips non-connector rows there. Hosted reads every source exactly as before (covered by a test asserting no source filter is applied). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test(entitlements): pin the early-grants wave on the self-host partition, fix stale docstrings Address the review round on the readGrants() merge resolution: - Add two recording-mock tests for getCompanyEntitlements with teamId (the dashboard layout path): on a self-host the early grants read must narrow to the connector keys and carry source = 'connector'; on hosted it reads every paid key with no source filter. The in-loop source check masked a lost narrowing, so this pins the query itself. - Move the getCompanyEntitlements docstring back above the function and replace "Self-hosted holds everything" with the actual partition. - keys.ts: say that lib/connect/instance arrives with stack PR #1748 so a reader on main does not chase a module that is not there yet. - DECISIONS.md: describe the InvoiceInboxWorkspace.tsx merge resolution accurately (set to the #1753 blob origin/main carried at push time, not the merged parent's) and note that stack children must merge this branch forward. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NdxH4DGnggvqHBCmif6B16 * chore(connector): merge origin/main, re-version connector migration to 20260831170000 Migration 20260820122000 predates versions already applied to prod (latest 20260831150000); renamed to keep Supabase branching history monotonic. Test reference updated. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(entitlements): own-credentials seam so self-hosts running their own EB/SKV are never connector-gated Forward-ports the connector-mode seam's own-credentials half from the instance-wiring layer: a self-host with its own Enable Banking or Skatteverket credentials holds that capability outright, exactly like every other local capability. Without this, upgrading an own-credentials self-host silently killed working bank sync and SKV integrations and showed a hosted subscription upsell whose remedy does not exist for a self-host. capability_blocked copy gains a self-host variant naming GNUBOK_CONNECTOR_KEY; stale all-on comments updated at the two gate call sites. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(entitlements): route the pending-op capability block through capabilityBlockedError for self-host copy Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb * fix(entitlements): PR #1747 review batch: connector-grant expiry CHECK, test cleanup, DECISIONS correction - Migration 20260831180000: CHECK (source <> 'connector' OR expires_at IS NOT NULL); connector grants are a short-lived offline cache, a NULL expiry would be a permanent unlock nothing revokes. pg test added. - beforeEach vi.clearAllMocks() in the three entitlements test files. - DECISIONS entry corrected: the trial seed DOES write trial rows for bank_sync/skatteverket; what it never writes is source='connector' or the connector-only keys. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UzNkSsR18pLFitJdYn8QEb --------- Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: Emil <emilmattsson14@gmail.com> |