Tracely-ai
Trace-native CI/CD for AI agents, production failures become regression tests that block the PR. Auto-detect, cluster, freeze into hermetic cases, replay in CI for $0.
Runssource: GitHubhomepagePythonMITcommit f332bc865d60
Python, MIT licensed. The project labels itself: agent, agent observability, ai agents, ci cd, clickhouse, evals, evaluation and llm.
Tracely-ai runs. An Argusic agent installed it in 12 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
All Python dependencies and frontend node_modules install, the backend API server starts on port 8000, the frontend builds with Next.js, the hermetic test suites (back-end, SDK, and frontend) pass 1478 tests total, and the Python codebase passes ruff lint.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 90 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.

What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- frontend vitest failed: ERR_REQUIRE_ESM from jsdom 30 on Node 183 minutes
- uv binary not on PATH after pip install1 minute
- pnpm global install blocked by filesystem permissions1 minute
- pytest module not found in PATH1 minute
- Install time
- 12 minutes
- Cold machine to finish
- 12 minutes
- Errors hit and fixed
- 4 hit, 4 fixed with no human help
- How the result was proved
- python3 -m uv run pytest -q backend/tests sdk/tests: 1174 passed, 34 skipped; cd frontend && npx pnpm test: 304 passed (37 files); cd frontend && npx pnpm build: all routes compiled; curl -s -w '%{http_code}' localhost:8000/health: 503 (expected without infra); python3 -m uv run ruff check .: all checks passed
- Model tokens used
- 54,473
- Exact commit tested
- f332bc865d60
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
Watch the session
The whole run, from a clean machine, at the speed it actually happened. Nothing is cut and nothing is sped up.
A recording of attempt 1, run . The verdict above comes from a different attempt, so read this as one session and not as the current result. The terminal recording it was made from is the evidence, stored with a sha256 fingerprint; the video is a rendering of it. Nothing loads from the video host until you press play.
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
54,473 All verified. Let me record the final assessment. Here is the complete summary of everything I observed and verified. **Install**: `uv sync --all-packages --all-extras` resolved and installed all Python packages (back… **Errors encountered and fixed**: 1. `uv` binary not on PATH — installed via `pip3 install uv --break-system-packages` 2. `pnpm` couldn't install globally (permissions) — used `npx pnpm` as a drop-in 3. Node 18 was too old for jsdom 30's ESM loading — downloaded Node 20, downgraded jsdom to `25.0.1` 4. `pytest` not directly on PATH — used `python3 -m uv run pytest` to run within venv **Tests**: Backend: 1008 passed, 34 skipped. SDK: 166 passed. Frontend: 304 passed (37 files). Tota… **Backend server**: Starts and listens on port 8000. Health endpoint returns 503 because ClickHouse… **Frontend build**: Next.js build succeeds — all routes compile, middleware passes type checks.
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
What did not, or is not known
- Took 12 minutes to install, slower than the median of the 140 comparable projects Argusic has measured.
Of the 141 agent projects Argusic has installed and timed, Tracely-ai was the 95th fastest to reach a running state, and 94 of 141 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
agentagent-observabilityai-agentsci-cdclickhouseevalsevaluationllmllm-as-judgellm-evaluationllm-observabilityllm-opsllmopsmcpmonitoringopentelemetrypythonself-hostedtracing
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/tracely-ai)Questions
- Does Tracely-ai run?
- Yes. Tracely-ai runs. Argusic installed and launched it on a clean machine in 12 minutes, hitting 4 errors on the way, and recorded the session.
- How did Argusic test Tracely-ai?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit f332bc865d60. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test Tracely-ai?
- The run that produced this verdict cost $0.05: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does Tracely-ai take to install?
- 12 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does Tracely-ai need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing Tracely-ai?
- 4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does Tracely-ai compare with the alternatives?
- Of the 141 agent projects Argusic has installed and timed, Tracely-ai was the 95th fastest to reach a running state, and 94 of 141 reached one at all.
- Where is the evidence for Tracely-ai?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.