feat(telemetry): project verify and outcome receipts

[solomonneas]

Closes #491

Parentage and envelope version

• A verify receipt named by worker-results.json shares the owning run's trace ID and uses the first valid projected worker span as its parent.
• An orphan verify receipt is a deterministic root span with parent_span_id: null. If separate runs claim the same receipt ID, the ownership is ambiguous and the receipt follows the same orphan rule.
• Outcome records inherit the run anchor resolved from an exact run or verify receipt evidence path.
• Both envelope schemas remain at v1. Receipt rows are additive, and literal success and failure snapshots pin the existing run-only output.

Attribute map

Receipt	Projected fields
Verify	executable basename, exit code, receipt status, duration, target repo, receipt ID, harness name, harness fingerprint
Outcome	artifact ID and kind, source, signal, current fingerprint-cohort score counts, current promotion status, target repo

OTel rows use gen_ai.operation.name=execute_tool, gen_ai.tool.name, and gen_ai.tool.call.id. OpenInference rows use openinference.span.kind=TOOL, tool.name, and tool.id. Brigade fields stay under brigade.*.

Only a canonical outcome ledger digest is exposed as brigade.receipt.id. Digestless or forged records get a deterministic correlation ID without claiming it is a receipt ID. Verify commands export the executable basename instead of raw arguments.

Tests

• Verify receipt inside a run in both formats, including trace and parent span
• Outcome receipt with run and verify evidence links in both formats
• Deterministic orphan and ambiguous-owner behavior
• Current fingerprint-cohort score and promotion attributes
• Canonical, forged, digestless, and duplicate outcome identities
• Receipt privacy and malformed-field handling
• Literal success and failure snapshots for unchanged run-only consumers

Verification

Command:

brigade work verify run --target . --command './scripts/verify' --capture brigade-work

• Receipt: 20260726-215908-work-verify-17a0b5
• Exit code: 0
• Suite: 4374 passed, 3 skipped
• Coverage: 82.81%
• MiseLedger evidence: f3e6699e9965e3c33ef2383b

Real linked projections

The source pair is run 20260726-issue491-implementation and verify receipt 20260726-205418-work-verify-d32500.

OTel GenAI:

{"attributes":{"brigade.receipt.id":"20260726-205418-work-verify-d32500","brigade.target.repo":"brigade-491-receipt-telemetry","brigade.verify.command.executable":"pytest","brigade.verify.exit_code":0,"brigade.verify.status":"completed","gen_ai.operation.name":"execute_tool","gen_ai.tool.call.id":"20260726-205418-work-verify-d32500:1","gen_ai.tool.name":"brigade.work.verify"},"duration_seconds":0.731811,"end_time":"2026-07-26T20:54:20.401969+00:00","error_type":null,"mapping":{"mapping_version":1,"name":"otel-genai","upstream_revision":"opentelemetry-semconv-1.43.0"},"name":"execute_tool brigade.work.verify","parent_span_id":"394b57b2349b7611","schema":"brigade.otel_genai_projection.v1","span_id":"03fd83e0ed1599a4","start_time":"2026-07-26T20:54:19.670019+00:00","status":"OK","trace_id":"3e1f5b04740e5bb51d2cc2fed3231dc4"}

OpenInference:

{"attributes":{"brigade.receipt.id":"20260726-205418-work-verify-d32500","brigade.target.repo":"brigade-491-receipt-telemetry","brigade.verify.command.executable":"pytest","brigade.verify.exit_code":0,"brigade.verify.status":"completed","openinference.span.kind":"TOOL","tool.id":"20260726-205418-work-verify-d32500:1","tool.name":"brigade.work.verify"},"duration_seconds":0.731811,"end_time":"2026-07-26T20:54:20.401969+00:00","error_type":null,"mapping":{"mapping_version":1,"name":"openinference","upstream_revision":"audited-2026-07-12"},"name":"brigade.work.verify","parent_span_id":"394b57b2349b7611","schema":"brigade.openinference_projection.v1","span_id":"03fd83e0ed1599a4","start_time":"2026-07-26T20:54:19.670019+00:00","status":"OK","trace_id":"3e1f5b04740e5bb51d2cc2fed3231dc4"}

[coderabbitai]

Important

Review skipped

Auto reviews are limited based on label configuration.

🏷️ Required labels (at least one) (1)

• coderabbit-review

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository: escoffier-labs/coderabbit/.coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 70d5de87-c5e4-497d-ab31-bf4d580a7747

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/491-receipt-telemetry

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}