agentleFS
tuanductrantuanductran/hr-skills/AGENTS.md

hr-skills

AGENTS.md58 starsChanged 2 months agoMITView on GitHub

hr-skills-monorepo is a Bun + Turborepo monorepo of domain-specific Agent Skills for HR and talent acquisition, distributed for Claude Code and claude.ai. Each skill lives at skills/hr-*/SKILL.md. The publishable hr-skills package provides the hr-skills npm executable; packages/hr-skills-build and packages/hr-skills-ref provide the TypeScript library surfaces, while packages/hr-skills-tsdoc owns multi-package TSDoc API generation. Generated artifacts — docs/engineering/skill-matrix.md, registry/skills.json, .claude-plugin/marketplace.json — are derived from skill frontmatter and must never be hand-edited; regenerate them with the corresponding bun run command instead. This file is…

The file

We split this AGENTS.md into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every other command — the full list, plus what each one does and when to use it — is documented once, in .agents/skills/turbo/SKILL.md and …

    line 32
  2. Very specific:

    When you add a new skill directory (for example skills/hr-new-skill/SKILL.md), run bun run sync first — it auto-discovers hr-* skill …

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 42,283 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

HR skills development guide

Project overview

hr-skills-monorepo is a Bun + Turborepo monorepo of domain-specific Agent Skills for HR and talent acquisition, distributed for Claude Code and claude.ai. Each skill lives at skills/hr-*/SKILL.md. The publishable hr-skills package provides the hr-skills npm executable; packages/hr-skills-build and packages/hr-skills-ref provide the TypeScript library surfaces, while packages/hr-skills-tsdoc owns multi-package TSDoc API generation. Generated artifacts — docs/engineering/skill-matrix.md, registry/skills.json, .claude-plugin/marketplace.json — are derived from skill frontmatter and must never be hand-edited; regenerate them with the corresponding bun run command instead.

This file is the canonical, tool-agnostic entry point for both human contributors and AI agents. The detailed, day-to-day workflow guidance lives in .agents/skills/ and .claude/ — see Where things live below.

Branch strategy

[!IMPORTANT] Never develop on the main branch.

  • main is the publishing branch and only contains released skills.
  • dev is the development branch where all feature work, fixes, and content updates should occur.
  • Always create pull requests against dev, not main.
BranchPurposeDirect commits
mainPublishing (npx hr-skills --help)Forbidden
devDevelopment, tests, experimentsVia PR only

Quick start

bun install          # Install all dependencies (run once, or after package changes)
npx hr-skills --help # Run the published CLI without a global install
bun run validate      # Validate all skill SKILL.md files
bun run test          # Run tests across workspace packages

Every other command — the full list, plus what each one does and when to use it — is documented once, in .agents/skills/turbo/SKILL.md and .agents/skills/bun/SKILL.md, not repeated here.

Where things live

This file stays intentionally short. Everything below is a workflow, not a one-time fact about the project — and workflows are owned by a single skill file under .agents/skills/, so they can be loaded on demand and updated in one place instead of drifting across copies.

TopicOwned by
Full bun run / turbo run command reference.agents/skills/turbo/SKILL.md, .agents/skills/bun/SKILL.md
Commit message format and Conventional Commits types.agents/skills/github-awesome-copilot-git-commit/SKILL.md
SKILL.md required structure, frontmatter, and pre-publish checklist.agents/skills/hr-skills-maintaining/SKILL.md
Markdown formatting, content-design rules, and the blank-line-before-list rule.agents/skills/markdown/SKILL.md
Maintaining the root SKILL.md router.agents/skills/hr-root-router-maintaining/SKILL.md
Security review of skill content.agents/skills/skill-vetter/SKILL.md
Whether a change needs a changeset.agents/skills/changeset/SKILL.md
TypeScript, Biome, and Valibot conventions in packages/*.agents/skills/typescript/SKILL.md, .agents/skills/biome/SKILL.md, .agents/skills/valibot/SKILL.md
CLI output patterns (@clack/prompts) in packages/hr-skills/src/cli/*.ts.agents/skills/clack/SKILL.md
Detecting and refactoring copy-paste duplication in packages/*.agents/skills/jscpd/SKILL.md, .agents/skills/dry-refactoring/SKILL.md

When you add a new skill directory (for example skills/hr-new-skill/SKILL.md), run bun run sync first — it auto-discovers hr-* skill directories from skills/ and updates .claude-plugin/marketplace.json. No manual edits needed.

Project structure

PathPurpose
skills/hr-*/SKILL.mdSource skill definitions consumed by Claude Code and claude.ai
skills/hr-*/content/Optional human-readable companion guidance for each HR skill domain
skills/hr-*/prompts/Optional reusable prompt libraries grouped by HR topic
skills/hr-*/examples/Optional practical end-to-end HR workflows and business scenarios
apps/web/Public Next.js documentation and product surfaces
playground/Experimental Next.js/Vite integration and smoke-test apps
docs/Skill format specification, architecture guidance, generated reports, and archived research
docs/engineering/skill-matrix.mdGenerated skill maturity snapshot — do not edit manually, run bun run matrix
docs/engineering/evaluation.mdEvaluation framework architecture, dataset format, and golden fixture workflow
packages/hr-skills-build/eval/datasets/Hand-authored evaluation datasets (planning scenarios)
packages/hr-skills-build/eval/golden/Committed golden fixtures — regenerate with bun run evaluate -- --update-golden from the repo root
docs/engineering/registry.mdSkill Registry architecture, schema, and extension guide
registry/skills.jsonGenerated machine-readable skill registry — do not edit manually, run bun run registry
.claude-plugin/marketplace.jsonGenerated marketplace metadata synced from skill frontmatter
packages/hr-skillsPublishable hr-skills executable package for npx/bunx and local CLI workflows
packages/hr-skills-buildBuild and maintenance tooling — validation, sync, registry/planner/runtime generation, and packaging
packages/hr-skills-refTypeScript library with explicit client and Bun/Node server surfaces for reading and validating skill files
packages/hr-skills-tsdocTSDoc-compatible API generator for all public package surfaces
docs/engineering/package-architecture.mdCanonical client/server package boundaries and import rules
.claude/rules/package-architecture.mdPath-scoped enforcement guidance for client/server imports

This repository uses Bun workspaces with Turborepo task orchestration; the packages above live under packages/*, and their build outputs are cached through Turborepo based on turbo.jsonc.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.