Production logs (2 months) show ~28 event_log inserts dying with TypeError: fetch failed. Root cause: the MCP server emits telemetry fire-and-forget and returns the JSON-RPC response immediately, so the insert races Vercel function suspension; supabase-js surfaces the dead fetch as a network error which was logged at error level. Two-part fix: - persistEvent (event-log-handler.ts) retries the insert once after 250ms when the error message contains "fetch failed" (network class only; constraint violations and other Postgres errors are never retried). On final failure, telemetry event types (mcp.*, agent.*) log at warn; business events (journal_entry.*, invoice.*, etc., which feed webhook delivery) stay at error. - The mcp-server telemetry emit sites (tool_called, tools_list_called, resource_read, next_hint_followed, skill_loaded, workflow_started, agent.feedback) now schedule the emit via after() from next/server, which keeps the function alive past the response until the emit settles. Falls back to plain fire-and-forget when no request scope exists (direct handler invocation in tests). Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
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.
additionalProperties: falseon everyinputSchema. Guarded bystrict-schemas.test.ts. Forces clear rejections on hallucinated fields instead of silent ignores.- Descriptions ≤ 280 chars. Guarded by
output-schema.test.ts. NoArgs:/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?. Thestaged: trueboolean is the explicit completion signal; agents must not infer completion from prose. Do NOT introduce a parallel{ success, shouldContinue, output }envelope. period_statusthreading: any tool that ties to a fiscal-period-bound date (categorize, mark paid, create voucher, correct/reverse entry, approve supplier invoice) passesdateForPeriodChecktostagePendingOperation. Response then includesperiod_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.tsTOOL_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 tolib/pending-operations/__tests__/voucher-executors.test.tsif the tool stages a newoperation_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 dispatchertool-result.ts:withNext(),toToolError()response helpersresources/: read-onlyAccounted://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 promptsskills/: domain-knowledge skill bodies served viagnubok_load_skill__tests__/: strictness guards + per-tool coverage