RECORD
THE PATH
NOT THE NOISE
Catch every time your AI coding agent touches auth, secrets, or skips a test, then turn the correction you made into a local regression eval.
Local-first and deterministic, with no LLM judge. TreeTrace reads the session transcript on your machine, flags every touch of auth, secrets, or tests and every risky command, and captures the human fix as a rerunnable eval. No upload. No telemetry. Record the path, not the noise.
Any AI agent or chat session, not only coding agents.
TreeTrace reads your coding and CLI agent sessions, and a plain User / Assistant chat transcript too. A regular ChatGPT, Claude, Gemini, or Grok conversation works the same way. No coding required.
- Claude Code
- Codex
- Cursor
- Copilot
- Gemini
- Grok
- ChatGPT
- Claude
- Gemini
- Grok
The 80-second tour.
Watch TreeTrace turn a raw agent session into a deterministic record, and the correction you made into a regression eval you can rerun.
Captions available in the player controls.
One local record. Three ways to read it.
Git history shows what changed. TreeTrace shows how the work actually got done, and keeps it as evidence you can replay, query, and hand off.
How it Works
Reconstruct a session's lineage from local transcripts: the root goal, every steer, the corrections that fixed it, and the dead-ends you abandoned, including the one that leaked a token.
Reconstruct a session 02 · What it RecordsWhat it Records
Every node captures the prompt, the edit, the tool call, files touched, refusals, and the outcome, written to your repo as an open, vendor-neutral schema your tools can read.
See the data model 03 · Use CasesUse Cases
A redacted, deterministic chain of custody for what an agent did, was corrected on, and was refused. One record that serves audit, onboarding, and prompting efficiency.
See where it earns its keepFrom raw transcript to structured record in one command.
Run it in any repo after an AI coding session. It reads local transcripts, never the network.
Discover
Claude Code sessions are found automatically from your local history. Codex, Cursor, Copilot, Gemini, Grok, and plain transcripts import with --from. Tool noise, retries, and "continue" nudges are filtered out.
Reconstruct
A fork-aware tree is derived from prompt topology and your text: the root goal, direction changes, corrections, abandoned branches, checkpoints, and the accepted path, with failure and refusal signals attached.
Export
Structured artifacts are written locally for humans, agents, CI, and eval harnesses. Every export passes a redaction gate that fails closed if a secret is detected, and a read-only treetrace mcp server hands the next agent its lessons.
One record, many uses.
The same local lineage answers three different questions, for three different people.
Engineers steering agents
See the cost of rework and where steering was needed. Make prompting more efficient by seeing what the agent kept getting wrong, and stop paying for the same mistake twice.
Teams & onboarding
Real corrections become regression evals and a memory pack for the next run, with no LLM judge. Hand off what went wrong so the next agent (or teammate) starts already knowing.
Audit & compliance
A redacted, signed-off record of what an agent did and was refused. Foundation being built
Auditable in one sitting. Apache-2.0.
Zero runtime dependencies, no telemetry, no network calls. The analysis layer is deterministic rules tuned to a published taxonomy and scored against a seeded benchmark with a blind holdout. Every result reproduced on committed code before it ships.
Validated against a seeded benchmark with a blind holdout.
The analysis layer is scored by a deterministic harness: exact-match rules against a published taxonomy, no LLM judge. The benchmark seeds 40 adversarial scenarios, each pairing a real signal with a benign distractor, so it measures precision and recall, not just coverage. A blind holdout is held out of development, so the headline number reflects generalization, not memorization.
Per-scenario macro-F1
Bars show the final-run macro-F1 for each named scenario; the 0.93 headline is measured on the held-out split. Every result is reproduced on committed code, and the full 166-test suite gates every change.
Deterministic, no LLM judge
The scorer is exact-match rules against a published taxonomy. Same input, same score, every run, with no model in the loop to grade the output, and no nondeterminism to explain away.
Generalizes, not memorizes
A blind holdout is kept out of development. It first exposed overfitting; the 0.93 headline is scored on that held-out split, so gains carry to sessions the detectors have never seen.
Validated by signal class
Corrections and declines, credential and security exposure, hallucinated file references, destructive actions, and lesson quality are each scored independently. Precision held or improved as false positives fell.
Trace your last AI coding session.
One command, in any repo. Nothing leaves your machine. Node 18+.
npx treetrace