Dev shares AGENTS.md rule: skip unit tests after coding, rely on E2E only
alexcovo_eth · x · 2026-09-21
- Developer anshnanda shares the testing rules he pins at the top of his AGENTS.md, prompted by a jab that Opus happily writes 10 unit tests checking whether a constant string contains substrings.
- His rules: never write unit tests after writing code; prefer E2E tests as the sole testing mechanism to verify complex features, ending each run with a verifiable, repeatable artifact.
- For isolated system tests, first enumerate all the ways the system could fail, then write the code.
- The workflow targets a common LLM failure mode: generating piles of low-value unit tests.
More from coding & agent
- AI Comes for the If Statement: Tomasz Tunguz on AI replacing hardcoded rule logic — hardimanjames · 2026-09-22
- Thorsten Ball: Code Review Will Die, and Unit Tests Might Follow — rseroter · 2026-09-22
- Tinfield 1 open-weight model claims to beat Claude Opus 4.8 on Terminal-Bench — victormustar · 2026-09-22
- Qwen Code Desktop v0.24.3 ships bwrap sandbox, DingTalk output and token budgets — github-actions[bot] · 2026-09-22
- Hazy Research: agents are replacing abstractions — CUDA DSLs are heading to retirement — ricklamers · 2026-09-22
- Musk reveals multi-agent setup: Grok Bot orchestrates Claude Code, Codex, hints at Grok 4.7 — elonmusk · 2026-09-22