Observed arrival · 2026-10-18
ProofRun: receipts for the tests an AI agent says it ran
An offline, open-source CLI records which checks ran against a particular code state and reports whether their results still apply.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Developers checking AI-generated code
- Worth noticing
- It compares actual argument arrays, so a similarly worded command cannot stand in for the declared check.
Field notes
The documented workflow includes a strict status mode that exits nonzero when any check is not PASS, suitable for a pre-commit hook or CI gate. The page says the included GitHub Action reruns checks against the PR head rather than trusting a receipt committed to the repository. The project explicitly excludes parsing test output, judging code quality, and auto-fixing.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue