● Covered by 1 source · 1 reportLow impact1 neutral
ProofRun Tool Provides Cryptographic Verification for AI Coding Agent Test Results
Key points
- ProofRun cryptographically verifies test execution against current code.
- It binds check results to a code fingerprint, marking them STALE if code changes.
- The tool operates offline and does not use AI for verification.
- It supports four statuses: PASS, FAIL, STALE, NOT RUN.
Addressing AI Agent Test Reporting
ProofRun is a new utility designed to provide cryptographic verification for test results reported by AI coding agents. The tool aims to solve the problem of AI agents potentially reporting outdated or inferred test statuses, rather than results from actual, recent executions. It ensures that when an AI agent claims "all tests pass," there is a verifiable record of that execution against the exact current codebase.
How ProofRun Works
ProofRun operates by binding every check result to a unique fingerprint of the code's state, which includes the git commit and a hash of all uncommitted changes. If any byte of the code changes after a test run, the previous result is automatically marked as STALE. This mechanism eliminates the need for manual checks on whether a reported "PASS" status is still valid.
The tool initiates a real subprocess to run tests and reads its exit code, rather than using AI for verification. It supports four distinct statuses: PASS, FAIL, STALE, and NOT RUN, each reflecting an observed execution or its absence. ProofRun does not parse test output, judge code quality, or auto-fix issues.
Key Features and Implementation
ProofRun is fully offline, with no network calls, telemetry, or accounts required. It compares exact argument arrays for commands, preventing similar-looking but functionally different commands from satisfying a check. The tool can be installed via a direct download, built from source, or using `go install`.
Users initialize ProofRun with `proofrun init`, run tests with `proofrun run test -- <command>`, and check status with `proofrun status`. A `--strict` flag ensures a non-zero exit code if any check is not PASS.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
GitHub cmc_internal/api GitHub github/collect GitHub yebiguo/ProofRun GitHub _private/browser GitHub get-started/accessibility GitHub open-source/sponsorsReporting from
ProofRun is a new tool that cryptographically verifies whether tests have actually run against the current code state when an AI coding agent reports "all tests pass." It addresses the issue of AI agents potentially reporting stale test results by binding check outcomes to a precise code fingerprint, marking results as STALE if the code changes. This tool provides developers with a reliable, offline mechanism to confirm the recency and validity of test executions reported by AI agents.