Assay: an open-source Claude Code skill that catches AI behavior regressions before commit
Professional-Run3614 · reddit · 2026-10-15
A developer open-sourced Assay, a Claude Code skill that catches AI behavior regressions passing tests can miss — e.g. a refund test still passes but the agent quietly stopped asking for approval.
Workflow:
- Claude Code changes a prompt, tool, or agent logic
- The Assay skill runs relevant AI regression tests
- Each case is compared against its last passing baseline, showing behavior diffs
- Claude Code investigates and proposes a fix before commit
Baselines never update automatically, so regressions can't become the new norm; the skill also forbids Claude Code from weakening or deleting tests to make them pass. Repeated runs help separate real regressions from flaky results. Supports pytest, Jest, and Vitest; MIT-licensed, installable via pip install assay-server pytest && assay init --claude-code.
More from coding & agent
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23
- Capy 0.5.0 lets Claude Pro/Max subscribers reuse their plan quota, adds OIDC cloud access — garrytan · 2026-10-23
- You.com CEO Richard Socher: legacy code is human slop, let AI clean it up — RichardSocher · 2026-10-23
- Let agents design a single-line glance component so you can track them from the sidebar — blixt · 2026-10-23