Write and audit AI coding agent instruction files for multi-tool repositories.
npx skills add https://github.com/tankpkg/packages --skill tank-agents-md
Please help me install this Agent Skill. Skill: @tank/agents-md Source: https://github.com/tankpkg/packages/tree/main/skills/agents-md Command: npx skills add https://github.com/tankpkg/packages --skill tank-agents-md
This Skill helps you create, audit, and manage AI coding agent instruction files without duplicating guidance across multiple tools or letting rules drift out of sync.
Use the agents-md skill to analyze your repository and draft a concise, cross-tool instruction strategy for AGENTS.md, Claude Code, Cursor, Copilot, Windsurf, Gemini, and OpenCode.
High-intent search queries and answers about installing and using this skill.
To manage AI coding agent instruction files without duplicating guidance, use a cross-tool instruction strategy with canonical source-of-truth patterns and minimal tool-specific compatibility bridges for AGENTS.md, CLAUDE.md, and Cursor rules.
The best way to structure instruction files in a monorepo is using layered instruction files with path-scoped overrides. This monorepo rule architecture applies precedence awareness to ensure package-scoped workflows remain accurate across tools.
Yes, this approach supports GitHub Copilot instructions, Windsurf rules, and Gemini CLI simultaneously. Cross-tool instruction design covers AGENTS.md, CLAUDE.md, Cursor, Aider, and OpenCode, applying frontmatter discovery metadata to maintain compatibility.
To migrate rules between AI coding agent formats safely, apply safety-aware rule writing practices. This involves writing concise, non-duplicative instructions and using cross-tool compatibility bridges to translate formats without losing context or precedence.
You should exclude duplicated guidance and redundant context to keep AI coding agent instruction files concise. The strategy emphasizes safety-aware rule writing, ensuring only minimal, necessary instructions are included for repository-wide and package-scoped workflows.
lrbjkAssess Unity projects for version, render pipeline, packages, asmdefs, and folder structure.
bahayonghangAudit and optimize repository-scoped Codex instruction chains in AGENTS.md files.
concordance-coDesign and edit pipelines_v2 workflows with runner specs and metadata.
euroconicCreate decoupled AI-native project folders with explicit agent rules.
