Agent Skill: Generate and maintain AGENTS.md files following the public agents.md convention. Use when creating AI agent documentation, onboarding guides, or standardizing agent patterns. By Netresearch.
Generates and maintains AGENTS.md files following the public agents.md convention. Use when creating AI agent documentation, onboarding guides, or standardizing agent patterns across Go, PHP, TypeScript, Python, and hybrid projects.
/plugin marketplace add netresearch/agents-skill/plugin install agents@netresearch-claude-code-marketplaceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
references/analysis.mdreferences/directory-coverage.mdreferences/examples/coding-agent-cli/AGENTS.mdreferences/examples/coding-agent-cli/scripts-AGENTS.mdreferences/examples/ldap-selfservice/AGENTS.mdreferences/examples/ldap-selfservice/internal-AGENTS.mdreferences/examples/ldap-selfservice/internal-web-AGENTS.mdreferences/examples/simple-ldap-go/AGENTS.mdreferences/examples/simple-ldap-go/examples-AGENTS.mdreferences/examples/t3x-rte-ckeditor-image/AGENTS.mdreferences/examples/t3x-rte-ckeditor-image/Classes-AGENTS.mdscripts/detect-project.shscripts/detect-scopes.shscripts/extract-commands.shscripts/generate-agents.shscripts/lib/template.shscripts/validate-structure.shtemplates/root-thin.mdtemplates/root-verbose.mdtemplates/scoped/backend-go.mdGenerate and maintain AGENTS.md files following the public agents.md convention.
Create AGENTS.md in ALL key directories, not just root:
| Directory | Purpose |
|---|---|
| Root | Precedence, architecture overview |
Classes/ or src/ | Source code patterns |
Configuration/ or config/ | Framework config |
Documentation/ or docs/ | Doc standards |
Resources/ or assets/ | Templates, assets |
Tests/ | Testing patterns |
See references/directory-coverage.md for full guidance.
# Generate for current project
scripts/generate-agents.sh . [--dry-run] [--verbose]
# Template styles
scripts/generate-agents.sh . --style=thin # Default, ~30 lines
scripts/generate-agents.sh . --style=verbose # ~100-200 lines
# Update existing / Validate
scripts/generate-agents.sh . --update
scripts/validate-structure.sh .
Root (thin, ~30 lines): Precedence statement, global rules, pre-commit checks, scope index
Scoped files (9 sections): Overview, Setup, Build/tests, Code style, Security, PR checklist, Examples, When stuck, House Rules
references/analysis.md - Analysis of 21 real AGENTS.md filesreferences/examples/ - Real-world AGENTS.md filesreferences/best-practices.md - Writing guideContributing: https://github.com/netresearch/agents-skill
This skill should be used when the user asks to "create a hookify rule", "write a hook rule", "configure hookify", "add a hookify rule", or needs guidance on hookify rule syntax and patterns.
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.