A Claude Code / Codex skill for designing and reviewing CLI tools that AI agents can call reliably — non-interactive mode, JSON output contracts, layered exit codes, dry-run, idempotency — while staying usable for humans.
1 file for Codex and OpenCode: agent-friendly-cli-skill AGENTS.md — 406 tokens loaded in every session.
AGENTS.md A 406 tok These files are x0c/agent-friendly-cli-skill's own configuration — they tell Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.