Files
accounted/lib/docs/content/connect-claude.ts
T
Jakob Wennberg ec27228a8e style: remove em/en dashes repo-wide, add CLAUDE.md rule against them (#890)
Em dashes (—) and en dashes (–) had spread across comments, docs, tests,
and a few UI strings, reading as AI-generated boilerplate rather than
house style. Replaced each with punctuation matching its context: colon
for explanatory clauses, comma for asides, plain hyphen for numeric/legal
ranges (e.g. "21-23§"), "to"/"till" for date ranges, parentheses for
paired-dash asides. messages/en.json and messages/sv.json were fixed by
hand together to keep sv/en in sync.

Left untouched where the dash is the functional subject rather than
decorative punctuation: date-range-parser.ts's separator regex,
charset-repair.ts's CP1252 byte-mapping table (and its test), the SIE
encoding mojibake docs, generic-csv.ts's minus-sign normalizer, the
agent system-prompt files that already instruct against em dashes, and
a golden iXBRL test fixture compared byte-for-byte.

Also fixes two bugs surfaced along the way: an off-by-one in
ApiKeysPanel's scope-label split (a leftover from an earlier partial
pass), and a charset-repair test that had lost the literal en-dash it
exists to verify.

Regenerated the agent atom seed migration (skills:generate) since 27
SKILL.md files changed. Added a CLAUDE.md rule against em/en dashes,
with an explicit carve-out for the functional-dash cases above.

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-04 15:58:06 +02:00

81 lines
5.8 KiB
TypeScript

export const CONNECT_CLAUDE_MD = `# Connect with Claude
> Talk to your bookkeeping. Connect Accounted to Claude (claude.ai, Claude Desktop, or Claude Code) and ask questions, categorise transactions, and prepare a momsdeklaration in plain language: every write still stages for your approval first.
Accounted ships an [MCP](https://modelcontextprotocol.io) server that exposes the full bookkeeping engine (90+ tools) to any MCP client. The endpoint is:
\`\`\`
https://app.gnubok.se/api/extensions/ext/mcp-server/mcp
\`\`\`
There are two ways to connect, depending on your client. Both reach the same tools and the same approval model: read tools answer immediately, write tools (categorise, mark paid, create voucher, year-end) **stage a pending operation** that you confirm in chat or in the **/pending** web UI before anything is booked.
## Path A: claude.ai / Claude Desktop custom connector (OAuth 2.1)
Best for most users. No API key to manage: you authorise Accounted the same way you'd authorise any OAuth app.
1. In **claude.ai** (Settings → Connectors) or **Claude Desktop** (Settings → Connectors → Add custom connector), choose **Add custom connector**.
2. Paste the connector URL:
\`\`\`
https://app.gnubok.se/api/extensions/ext/mcp-server/mcp?client=claude-connector
\`\`\`
_(The "?client=claude-connector" suffix is telemetry-only: it lets Accounted see you connected via claude.ai/Desktop. It changes nothing about behaviour or scopes; drop the query string if you prefer.)_
3. Claude opens the Accounted OAuth 2.1 consent screen. Sign in and pick the company you want Claude to act on.
4. On the consent screen you grant **read-only scopes by default** (list invoices, read reports, compute VAT). Write scopes (create invoice, categorise, book vouchers, run year-end) are **listed separately and must be ticked explicitly**: leave them unchecked for a read-only review session.
5. Approve. Claude now lists the Accounted tools and you can start asking questions.
Because the consent is per-company and scoped, you can connect a read-only key for a reviewer and a separate write-enabled connection for day-to-day bookkeeping.
## Path B: \`npx gnubok-mcp\` with an API key (stdio bridge)
Best for Claude Desktop on a machine where you'd rather use a long-lived API key than the OAuth flow, or for scripting.
1. Mint an API key in the Accounted dashboard at **/settings/api**. Use a \`gnubok_sk_test_*\` key against the sandbox while you evaluate; switch to \`gnubok_sk_live_*\` for real data.
2. Add the stdio bridge to your \`claude_desktop_config.json\`:
\`\`\`json
{
"mcpServers": {
"gnubok": {
"command": "npx",
"args": ["gnubok-mcp"],
"env": {
"GNUBOK_API_KEY": "gnubok_sk_test_...",
"GNUBOK_CLIENT": "claude-desktop"
}
}
}
}
\`\`\`
3. Restart Claude Desktop. The bridge proxies stdio JSON-RPC to the hosted MCP endpoint over HTTPS; the key carries the scopes you granted it at mint time.
The key's scopes gate exactly which tools are callable: a key without write scopes can read reports and ledgers but cannot stage a booking.
## Try these prompts
All three run against the deterministic sandbox seed (use a \`gnubok_sk_test_*\` key or pick the sandbox company on the OAuth consent screen). They exercise the read path end-to-end without booking anything.
1. **"Show my uncategorized bank transactions and suggest categories."**
Claude calls \`gnubok_list_uncategorized_transactions\` then \`gnubok_suggest_categories\` and walks you through the proposals. Approving one stages a \`gnubok_categorize_transaction\` pending operation: nothing is booked until you confirm.
2. **"Which invoices are overdue?"**
Claude calls \`gnubok_get_ar_ledger\` (kundreskontra) and lists outstanding customer invoices with aging.
3. **"Compute my VAT report for this quarter and tell me if I can close it."**
Claude calls \`gnubok_get_vat_report\` for the momsdeklaration rutor, then \`gnubok_vat_close_check\` to scan for blockers (uncategorised rows, unapproved supplier invoices, missing receipts on expenses ≥ 4 000 kr: the tool's high-value heuristic; BFL requires underlag for every affärshändelse regardless of amount) and reports \`ready_to_close\`.
## 10-minute reviewer test
A quick end-to-end pass to confirm the connection works before you trust it with real data. Run the steps in order; each lists what you do and what you should see.
1. **Connect.** Use Path A (read-only scopes only) or Path B with a \`gnubok_sk_test_*\` key. → Claude lists the Accounted tools (titles like *List Uncategorized Transactions*, *VAT Declaration (Momsdeklaration)*).
2. **Confirm the company.** Ask *"Which company am I connected to?"* → Claude names the sandbox company (e.g. **Sandlådan Konsult**).
3. **Run prompt 1** (*uncategorized + suggest categories*). → A list of uncategorised rows plus category suggestions; no booking happens.
4. **Run prompt 2** (*overdue invoices*). → At least one overdue customer invoice with aging.
5. **Run prompt 3** (*VAT report + can I close*). → Momsdeklaration rutor returned; \`vat_close_check\` reports a **non-empty blocker list** (uncategorised transactions, an unapproved leverantörsfaktura, and a high-value business expense without a receipt).
6. **Stage a write.** Ask Claude to categorise one transaction. → Claude stages a pending operation and asks you to confirm: the booking does **not** post until you approve in chat or at **/pending**.
If every step matches, the connector is wired correctly and the approval model is enforced.
## Support
Stuck connecting, or seeing an unexpected blocker? Use the in-app support form at **/help**: it routes straight to the product team with your company context attached. Include the client (claude.ai / Desktop / Code), the path you used (A or B), and the tool name from any error message.
`