jwuthri-cloud/Tracely-ai
Trace-native CI/CD for AI agents: grade production traces, cluster failures, freeze them into replayable regression tests, and block PRs that regress. Self-hostable.
Older snapshot: check the original source for current information.
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
Trace-native CI/CD for AI agents. Tracely grades every agent trace as it lands, clusters the failures into issues, freezes the bad runs into hermetic replayable cases, blocks the pull request that would ship them again — and tells you the moment any of it happens. production trace → failure detection → regression test → CI gate → alert Website · Docs · Product guide · Agent skill · Guided tour · 2-min demo · Design dossier [](https://github.com/jwuthri-cloud/Tracely-ai/actions/workflows/ci.yml) [](https://pypi.org/project/tracelyai-sdk/) [](https://pypi.org/project/tracelyai-sdk/) [](LICENSE) [](https://github.com/jwuthri-cloud/Tracely-ai/stargazers) [](https://mcpvault.io/servers/tracely/health?utmsource=externalbadge&utmmedium=referral&utmcampaign=mcphealthreport) Self-host the whole stack in one click — API, worker, UI, Postgres, ClickHouse, Redis and MinIO: [](https://railway.com/deploy/n5nLE?referralCode=WCq5Cn&utmmedium=integration&utmsource=template&utmcampaign=generic) [](https://app.railyard.run/deploy?repo=https://github.com/jwuthri-cloud/Tracely-ai) One workspace: what ran, what broke, what is already pinned as a test. Because observability stops at the dashboard. You can see that your agent broke — then what? Every eval tool asks you to hand-author a dataset: sit down, invent questions, write ideal answers, keep them current as the…
Read the full README ↗ · Preview checked 2026-11-01T11:52:15.177Z
Inside the original README — Document outline
- Tracely
- Production failures become regression tests.
- Why another agent-observability tool?
- The loop
- 1 · Observe — every run, hierarchically
- 2 · Detect & triage — failures group into issues
- 3 · Test — freeze the failure
- 4 · Ship — block the PR
- 5 · Tell me — alerts as a flow
- Plus — trends, cross-metric analysis, judge calibration
- Everything in the box
- Quickstart
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: MIT. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
Trace-native CI/CD for AI agents: grade production traces, cluster failures, freeze them into replayable regression tests, and block PRs that regress. Self-hostable.
Repository facts
- Owner
- jwuthri-cloud
- Primary language
- Python
- Stars
- 975
- Forks
- 140
- Open issues + pull requests
- 19
- License
- MIT
- Archived
- No
- Default branch
- main
- Created
- 2026-10-11T15:38:03.000Z
- Last push
- 2026-10-15T23:52:25.000Z
Topics and intended use
Owner-supplied topics: agent-evaluation, ai-agents, ai-testing, ci-cd, clickhouse, developer-tools, llm-evaluation, llm-observability, llmops, mcp, observability, python, regression-testing, self-hosted, tracing
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-10-18T12:10:07.336Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related repositories
EbookFoundation/free-programming-books
donnemartin/system-design-primer