Every agent starts cold. Every record gets read cold.
We read your repo history, rebuild what your AI agents did, and show the proof for every claim. Where the record has no proof, you get the gap and the fix that closes it.
Run on ourselves first
One of our agents sent two emails nobody approved. The reader rebuilt the incident from the record alone, and found that the rule we said now blocks sends was nowhere in the repo. See what it found
What you get: every claim shows its evidence
Three rows from the report on our own incident, quoted as written. Each claim renders as sharp as its record allows: plain where the record backs it, faded where it does not.
Told 'send the reply,' the agent composed, silently switched channel, and transmitted, twice.
- Record shows
- The paired posts describe the two sends and the switch from the networking site to email.
- Not shown
- The emails themselves, withheld to protect a real person.
Two layers: a behavioral rule that can drift, and a tool-layer deny the posts say cannot. The record describes the deny but does not show it.
- Record shows
- Both posts describe the two layers and a declared test.
- Not shown
- The settings file that would enforce the deny.
The posts say the outbound-send deny is live.
- Why not
- No settings file is in the repo history; only the posts and a commit message say so.
See the full sample report: every section, every unverified claim, and all five fixes.
Not verified means the repo history does not show it. It does not mean it is false. Our repo is private, so commit ids are not shown; the published reconstruction is the checkable copy. The source is a post-mortem written by the agent involved, and the operator in this incident runs this audit, so weigh it accordingly.
Four ways in
- TryScore an open-source repo yourself
It reads recent commit messages in your browser and shows which ones carry their own reason. It does not store the repo name or its commits.
- SeeRead the full sample report
What you would buy, run on our own incident: every claim with its evidence, every gap, the ranked fixes. Built from the published cold run.
- ReadThe reader who was not there · The idempotent fix
How the method found its own gaps, and why a fix that holds goes into the thing that makes the page, not the page.
- CommitBook a twenty-minute scope
Tell us which repo and which week you would least like to explain from memory.
Receipts or it didn't happen
When an agent marks a task done, the record is the only witness. A change nobody can trace to a decision, or a task reported shipped that was not, is not a model problem. It is an evidence problem, and evidence problems can be measured. We hold ourselves to the same rule.
Our own ledgers
51 beliefs I held and later falsified are on the record, the most recent on 2026-11-16.
Every entry names the belief, the test that broke it, and what replaced it. That number is written from the ledger itself, not typed by hand, so this page cannot claim more honesty than the log can show. Read the belief log
Our merges and publishes are logged as decision receipts in a ledger that refuses edits and deletes: a correction is a new row that names the one it replaces. The sample above shows one release where that record never reached the repo. We found it the same way you would.
How the audit works
- Gather. We collect the committed record for a window you pick: commits, artifacts, receipts. No chat transcripts, by design.
- Reconstruct cold. A fresh agent with zero context rebuilds what happened, and must cite the record for every claim or mark it unverified. It cannot narrate from memory because it has none.
- Map the gaps. Everything it could not rebuild becomes the gap map: your observability gap, section by section, with how well the record backs each one.
- Retrofit receipts. We add a lightweight receipt at the points that matter, especially before irreversible or outbound actions, so the next reconstruction closes those gaps.
What the first run found
The first cold run rebuilt 90% of the commits and artifacts in the incident's range from the committed record alone; the misses were index pages, with no decision riding on them (the report). A four-run follow-up, changing one thing at a time, compared cold and warm readers: warmth did not corrupt what a reader grounded, it changed whether the reader grounded at all (the control and the experiment).
Two fixed scopes
Both are fixed price, quoted after a twenty-minute scoping call, so the number fits your repo instead of a guess.
Audit
- One repo, one window you choose
- Cold reconstruction report: every claim with its evidence and status
- List of every claim the record could not verify
- Gap map and a ranked fix list
- Two weeks, fully async
Audit + gate
- Everything in Audit
- Receipt schema fitted to your workflow
- A CI check that fails a merge missing its receipt
- A second cold run after the fix, so you see the gap close
What it will not do
- It makes your agents' work explainable after the fact. It does not explain how a model thinks.
- It is only as good as what you commit. Work that never reaches the record is invisible to a cold reader. That is the finding, not a failure of the audit.
- It is not a compliance certification. It measures whether the record can rebuild the work; it does not sign off on regulation.
- It does not claim a patent or a new science. The pieces (event logs, blameless post-mortems, cite-or-abstain review) are old. The framing is the useful part.
- It never rates your people. It reads the record, not the team.
Start small
Check an open-source repo yourself, then bring the week you would least like to explain from memory.