
Create Agents Md
- 285 installs
- 20 repo stars
- Updated March 21, 2026
- siviter-xyz/dot-agent
create-agents-md is an agent skill that generates AGENTS.md files with project-specific inline rules, CLI commands, and safety boundaries for developers who want committed repo guidance for coding agents.
About
create-agents-md is a siviter-xyz/dot-agent skill that guides creation of AGENTS.md files for small, project-specific instructions committed inside repositories. It supports a root AGENTS.md for whole-project commands, tech stack, testing, code style, architecture, and safety boundaries, plus nested subdirectory AGENTS.md files for monorepo packages, services, or test trees where local rules diverge. Developers reach for create-agents-md when adding folder-scoped agent rules that should live in git rather than reusable cross-project skills. The skill recommends short concrete references, exact CLI invocations, links to canonical example files, and keeping files focused instead of long prose documentation.
- create-agents-md
Create Agents Md by the numbers
- 285 all-time installs (skills.sh)
- Ranked #1,388 of 4,347 Backend & APIs skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/siviter-xyz/dot-agent --skill create-agents-mdAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 285 |
|---|---|
| repo stars | ★ 20 |
| Last updated | March 21, 2026 |
| Repository | siviter-xyz/dot-agent ↗ |
How do you write an AGENTS.md file for a repository?
Use create-agents-md for development tasks
Who is it for?
Developers onboarding coding agents to a repository who need committed, scoped inline instructions instead of a reusable cross-project skill.
Skip if: Skip create-agents-md when the knowledge should be a reusable skill with scripts and references shared across multiple repositories.
When should I use this skill?
Adding small project-specific agent instructions, monorepo package rules, test-directory guidance, or folder-scoped conventions to commit in-repo.
What you get
Committed AGENTS.md with setup commands, code style, testing instructions, and agent safety boundaries
- AGENTS.md file
Files
Create AGENTS.md
Guide for creating AGENTS.md files for project-specific inline rules.
When to Use AGENTS.md
- Small, project-specific instructions that should be committed in the repo
- Folder-scoped rules for specific directories
- Package-specific instructions in monorepos
- Test-specific guidance in test directories
When NOT to Use AGENTS.md
- Reusable knowledge across projects → Use skills
- Large documentation → Use skills with references
- Complex workflows → Use skills with scripts
AGENTS.md Structure
AGENTS.md is a simple markdown file without metadata:
# Project Instructions
## Code Style
- Use TypeScript for all new files
- Prefer functional components in React
- Use snake_case for database columns
## Architecture
- Follow the repository pattern
- Keep business logic in service layersLocation
- Project root:
AGENTS.md– Primary, inline instructions and references for the whole project (commands, tech stack, testing, code style, architecture, safety boundaries). - Subdirectories:
subdirectory/AGENTS.md– Folder- or package-scoped instructions when local behavior meaningfully diverges from the root (e.g., a specific package, service, or test tree). - Nested support: Agents typically combine instructions from the closest
AGENTS.mdwith parent ones; keep root general and use nestedAGENTS.mdonly where you truly need more specific rules.
Best Practices
- Keep AGENTS.md files small and focused
- Use for project-specific conventions
- Prefer short, concrete references over long prose:
- Link to project docs, specs, and runbooks
- Point to example files or directories (e.g.,
see src/api/users.ts for canonical pattern) - Include the most important commands with exact CLI invocations
- Reference existing code examples when possible
- Update as project evolves
References
For detailed best practices, see references/best-practices.md.
AGENTS.md Best Practices
Size Guidelines
- Keep AGENTS.md files under 200 lines when possible
- Focus on essential project-specific instructions
- Move detailed documentation to skills or project docs
Content Guidelines
What to Include
- Project-specific coding conventions
- Architecture decisions unique to this project
- Testing patterns for this codebase
- Deployment procedures
- Environment setup
What to Avoid
- General programming advice (use skills instead)
- Framework documentation (reference official docs)
- Long examples (reference code files)
- Duplicate information from skills
Organization
Project Root AGENTS.md
- Single source of truth for global project conventions
- Overall architecture and key domain concepts
- Primary commands: build, test, lint, typecheck, format
- Tech stack details (framework versions, package managers)
- Safety boundaries (what the agent must not touch: secrets, vendor dirs, prod configs)
Folder-Scoped AGENTS.md
- Only add when a directory/package has meaningfully different rules than the root
- Directory-specific rules (e.g.,
apps/web/,packages/api/,infra/) - Package-specific instructions (in monorepos)
- Test-specific guidance (e.g., how to run slow/integration tests locally)
- Avoid duplicating root content; reference root conventions and add only the deltas
Examples
Good: Focused and Specific
# API Package Instructions
- All endpoints in `src/routes/`
- Use Zod for validation
- Return JSON with `{ success: boolean, data?: T, error?: string }`
- See `src/routes/users.ts` for exampleBad: Too Generic
# General Programming
- Write good code
- Use best practices
- Follow SOLID principles
- [500 lines of generic advice]Additional Online Guidance
These practices align with the broader AGENTS.md ecosystem:
- Use a root AGENTS.md as the main “instruction budget” for the project, then add nested AGENTS.md files only where local context really matters.
- Prefer concrete examples and commands over long prose; show exactly how to run tests or format code.
- Clearly define permissions and boundaries (what the agent can always do vs. what requires explicit user confirmation).
- Avoid auto-generating AGENTS.md from templates or scripts; keep it curated and project-specific instead.
Related skills
How it compares
Pick create-agents-md over skill-creator when the goal is per-repo inline agent rules rather than a portable reusable skill package.
FAQ
Where should AGENTS.md files live?
create-agents-md places a root AGENTS.md at the project root for whole-repo rules and optional nested AGENTS.md files in subdirectories when package, service, or test behavior diverges from root conventions.
When should developers use AGENTS.md instead of a skill?
create-agents-md recommends AGENTS.md for small committed project-specific instructions and skills for reusable knowledge, large documentation, or complex workflows with scripts across multiple repositories.