computer-playbook/docs/contributing/tools/agents.md

Back to CONTRIBUTING hub

Agent Guidelines for humans

AI assistants are welcome, but they must follow the same workflow as humans.

This page complements AGENTS.md. Use it as the agent-focused companion for contributor workflow, testing, review, and the code and development guides.

AGENTS.md is the repository-wide source of truth for agent-specific behavior.

Additional agent-specific rules:

• Do not invent commands, workflows, or files that do not exist.
• Never expose secrets in prompts, code, screenshots, or logs.
• Treat AI output as a draft that must be reviewed for wrong assumptions, duplicate logic, and security mistakes.
• Keep agent instructions consistent: AGENTS.md is the repository-wide source of truth, while CLAUDE.md and GEMINI.md are tool-specific additions.
• Read AGENTS.md before making changes.
• Keep explanations simple and explicit.
• State assumptions clearly.

More information here.