Files
accounted/extensions/general/mcp-server
Jakob WennbergandClaude Fable 5 11126d6d56 feat(dimensions): PR3 tagging — voucher-form pickers, MCP dimension tools with resolve-don't-select, engine soft validation (#859)
Phase 3 of dev_docs/dimensions_implementation_plan.md. Companies with
dimensions_enabled=false see zero change; existing free-text API writers keep
working (validation is toggle-governed).

Engine (soft validation):
- validateEntryDimensions() in dimension-resolver: zero queries for untagged
  entries; toggle off → passthrough; toggle on → one settings fetch + two
  registry queries, rejects unknown dims/codes and archived values with
  Swedish per-code messages (DimensionValidationError, 400, details.issues).
  Wired into createDraftEntry + updateDraftEntry before any insert; reversal/
  storno paths untouched (verbatim copies). Fails open on transient registry
  errors — soft validation must never block bookkeeping.

MCP (agent write path):
- New tools: gnubok_list_dimensions, gnubok_list_dimension_values (fuse.js
  fuzzy), gnubok_create_dimension_value (STAGED via pending_operations —
  agents never silently mint reporting values; new op type + CHECK migration
  + executor with duplicate-idempotency).
- create_voucher/correct_entry: per-line dimensions bag + default_dimensions,
  resolve-don't-select server-side (code OR natural-language name; exact →
  fuzzy ≤0.30 with ≥0.15 runner-up margin; non-exact resolutions echoed with
  confidence; ambiguous → ranked candidates, no auto-create).
- gnubok_get_agent_briefing gains a dimensions block (enabled, dims, top
  values) — omitted when registry empty.
- TOOL_SCOPE_MAP entries; risk tier low for staged value creation.

UI:
- JournalEntryForm (manual voucher + TransactionBookingDialog embed): header
  "+ Kostnadsställe/Projekt" progressive disclosure (gäller alla rader with
  documented inheritance rule) + per-row tag popover + compact KS·PR badges;
  gated on dimensions_enabled.
- Voucher detail: display-only dimension badges with registry-name resolution.
- EditDraftEntryDialog carries line dimensions so editing a draft no longer
  strips tags.

categorize/bulk_book dims deferred to PR7 (needs the bulk_book RPC migration).

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 13:36:45 +02:00
..
2026-05-04 11:12:29 +02:00

Accounted MCP server

JSON-RPC 2.0 server exposing the Accounted bookkeeping engine to MCP clients (Claude Desktop, Claude Code, etc.). Endpoint: /api/extensions/ext/mcp-server/mcp. OAuth and stdio bridge live alongside the API surface — see app/api/mcp-oauth/ and packages/gnubok-mcp/.

Tool authoring contract

Enforced by tests in __tests__/ — these are not style preferences, they're guard rails.

  1. additionalProperties: false on every inputSchema. Guarded by strict-schemas.test.ts. Forces clear rejections on hallucinated fields instead of silent ignores.
  2. Descriptions ≤ 280 chars. Guarded by output-schema.test.ts. No Args: / Returns: / Examples: prose — those belong in JSON Schema. Use agent-native hints ("Use to…", "Call X first", "HIGH risk").
  3. Staged-operation envelope for write tools — outputSchema: STAGED_OPERATION_SCHEMA (server.ts). Fields: staged, risk_level, actor, message, preview, period_status?, next?. The staged: true boolean is the explicit completion signal; agents must not infer completion from prose. Do NOT introduce a parallel { success, shouldContinue, output } envelope.
  4. period_status threading — any tool that ties to a fiscal-period-bound date (categorize, mark paid, create voucher, correct/reverse entry, approve supplier invoice) passes dateForPeriodCheck to stagePendingOperation. Response then includes period_status: { period_id, status: open|locked|closed, lock_date } so widgets and agents disable writes without round-trips.
  5. Scope mapping — every new tool needs an entry in lib/auth/api-keys.ts TOOL_SCOPE_MAP. Missing entries default to deny.
  6. Tests for new write tools — add staging-gate coverage to __tests__/voucher-tools.test.ts (or a sibling) plus executor coverage to lib/pending-operations/__tests__/voucher-executors.test.ts if the tool stages a new operation_type.

Determinism / cache stability

Tool definitions (name, description, inputSchema, outputSchema, annotations) are declared as static object literals at module load — no timestamps, no UUIDs, no Date/Math.random in the definition layer. This makes the tools/list JSON payload byte-stable across requests, which lets agent-side prompt caches stay warm. Do not introduce per-request non-determinism into the definitions block. Anything time-bound or random belongs inside execute().

For internal Anthropic API usage (today only extensions/general/invoice-inbox/lib/extract-invoice-fields.ts): annotate stable prefixes with cache_control: { type: 'ephemeral' } and log usage.cache_read_input_tokens for hit-ratio observability. The 1h TTL from the agent-native API plan (item 10) requires the direct Anthropic API; Accounted's Bedrock path defaults to a shorter TTL.

Payload-size watchdog

payload-size.bench.test.ts enforces a tools/list JSON payload ceiling (currently 36,000 tokens — bumped from 32,000 when top-level Tool.title landed on all tools for Claude Connectors Directory readiness). If the test fires, the right answer is rarely "raise the ceiling" — instead, trim descriptions or leverage gnubok_search_tools (already deployed; tool definitions can defer to it for discovery rather than enumerating in tools/list).

Where things live

  • server.ts — the tools array + JSON-RPC dispatcher
  • tool-result.ts — withNext(), toToolError() response helpers
  • resources/ — read-only Accounted:// URIs (active company, period, recent activity, capabilities, attention items, voucher gaps, chart of accounts, VAT treatments)
  • widgets/ — inline HTML widgets (receipt-matcher, vat-review)
  • prompts/ — slash-command-style prompts
  • skills/ — domain-knowledge skill bodies served via gnubok_load_skill
  • __tests__/ — strictness guards + per-tool coverage