Files

T

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>

2026-09-01 21:30:37 +02:00

__tests__

perf(mcp): trim the one schema 58 catalogued tools share, reclaiming 2 552 tokens (#2123 )

2026-09-01 21:30:37 +02:00

prompts

feat(bank): expose bank-reported balance (booked + available) in UI, reconciliation, MCP and v1 API (#2118 )

2026-09-01 16:16:29 +02:00

resources

feat(reconciliation): match one bank event to several verifikationer (1:N) (#1553 ) (#2029 )

2026-08-30 11:56:15 +02:00

skills

feat(transactions): steer private marking in locked periods to ignore, with v1 and MCP ignore verbs (#1661 ) (#2031 )

2026-08-31 08:39:04 +01:00

widgets

fix(bank): 90-day lookback is the default: long PSD2 requests make some banks kill the session (#1964 )

2026-08-26 22:36:06 +02:00

arg-guard.ts

feat(mcp): worked examples on five high-traffic tools, and in the error that rejects a call (#2100 )

2026-09-01 20:18:19 +02:00

company-routing.ts

feat(billing): multi-user paywall: multi_user capability, 20-day grace, owner-only dormancy (#2099 )

2026-09-01 11:29:12 +02:00

dimensions.ts

style: remove em/en dashes repo-wide, add CLAUDE.md rule against them (#890 )

2026-07-04 15:58:06 +02:00

index.ts

fix(mcp): honest tool text and build-derived server version (#1923 )

2026-08-26 13:35:15 +02:00

manifest.json

Bug/invalid imports (#1146 )

2026-07-24 15:03:50 +02:00

origin-guard.ts

style: remove em/en dashes repo-wide, add CLAUDE.md rule against them (#890 )

2026-07-04 15:58:06 +02:00

public-tools.ts

feat(mcp): lazy authentication so a client can connect before an account exists (#1814 PR 2) (#1892 )

2026-08-25 11:38:29 +02:00

README.md

fix(mcp): promote gnubok_update_customer to the default tools/list catalog (#1986 )

2026-08-27 22:28:14 +02:00

recommended-tools.ts

feat(transactions): steer private marking in locked periods to ignore, with v1 and MCP ignore verbs (#1661 ) (#2031 )

2026-08-31 08:39:04 +01:00

server.ts

perf(mcp): trim the one schema 58 catalogued tools share, reclaiming 2 552 tokens (#2123 )

2026-09-01 21:30:37 +02:00

staging-pii-guard.ts

fix(customers): personnummer via MCP lands in personal_number, masked everywhere; MCP payment terms follow settings (#1788 )

2026-08-21 18:32:17 +02:00

supplier-candidates.ts

fix(inbox): match suppliers on VAT number, not just org-nr and name (#1630 )

2026-08-17 14:52:50 +02:00

tasks.ts

feat(mcp): Tasks extension (io.modelcontextprotocol/tasks) (#1283 )

2026-07-29 18:15:06 +02:00

tool-namespace.ts

Bug/invalid imports (#1146 )

2026-07-24 15:03:50 +02:00

tool-result.ts

style: remove em/en dashes repo-wide, add CLAUDE.md rule against them (#890 )

2026-07-04 15:58:06 +02:00

README.md

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. Add ?tool_namespace=accounted for the Accounted tool names. Requests without it retain the legacy Gnubok namespace. OAuth and stdio bridges live alongside the API surface: see app/api/mcp-oauth/, packages/accounted-mcp/, and the compatibility package in packages/gnubok-mcp/.

Tool authoring contract

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

additionalProperties: false on every inputSchema. Guarded by strict-schemas.test.ts. Forces clear rejections on hallucinated fields instead of silent ignores.
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").
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.
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.
Scope mapping: every new tool needs an entry in lib/auth/api-keys.ts TOOL_SCOPE_MAP. Missing entries default to deny.
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 (the SDK is called from lib/ai/provider.ts and lib/ai/services/anthropic-family.ts; features such as extensions/general/invoice-inbox/lib/extract-invoice-fields.ts go through getAiService() in lib/ai rather than the SDK directly): 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. If the test fires, the right answer is rarely "raise the ceiling". Instead, trim descriptions or set specialized wide tools to catalogVisibility: 'search'. Those tools remain discoverable with full schemas through gnubok_search_tools and callable through tools/call on the wire without bloating the default catalog. Claude.ai only calls tools present in tools/list, so a tool that a user or a skill must call directly stays in the default catalog.

Where things live

server.ts: the tools array + JSON-RPC dispatcher
tool-result.ts: withNext(), toToolError() response helpers
resources/: read-only Accounted:// URIs, registered in resources/index.ts: company/current, period/active, recent-activity, capabilities, attention, chart-of-accounts, settings/vat-treatments, booking-templates, ledger/context, reconciliation/summary
widgets/: inline HTML widgets (receipt-matcher, vat-review, pending-operations)
prompts/: slash-command-style prompts
skills/: domain-knowledge skill bodies served via gnubok_load_skill
public-tools.ts: lazy authentication (issue #1814). ANONYMOUS_METHODS (initialize, ping, tools/prompts/resources listing) and the three PUBLIC_TOOLS (gnubok_search_tools, gnubok_list_skills, gnubok_load_skill) answer without credentials, rate-limited per truncated IP; every other tools/call gets a transport-level 401 + WWW-Authenticate from handleMcpRequest in server.ts, which is the challenge clients turn into their Connect prompt
tasks.ts: MCP Tasks extension (io.modelcontextprotocol/tasks): durable handles for long-running tool calls, rows in mcp_tasks (service-role writes only)
origin-guard.ts: Origin-header validation on the Streamable HTTP endpoint (DNS-rebinding defence required by the MCP spec)
staging-pii-guard.ts: refuses a plaintext personnummer in staged pending_operations params/preview, so every staging tool inherits the encrypt-at-staging rule
tool-namespace.ts: ?tool_namespace=accounted handling (accounted_* aliases for the canonical gnubok_* ids)
__tests__/: strictness guards + per-tool coverage