agentleFS
tuanductrantuanductran/hr-skills/.agents/skills/hr-skills-maintaining/SKILL.md

hr-skills-maintaining

Skill58 starsChanged 2 months agoMITView on GitHub

Repository-level skill describing metadata, directory structure, workflows, and conventions for the hr-skills monorepo. Use this skill when validating, creating, editing, or maintaining HR skill packages, including SKILL.md, content/, prompts/, and examples/.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 10 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    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. Distinctive:

    Verify directory structure and file organization

    line 83
  2. Distinctive:

    Verify internal relative links

    line 86
  3. Distinctive:

    Keep repository documentation synchronized with repository changes

    line 91
  4. Very specific:

    Always include a SKILL.md in every skill package.

    line 137
  5. Very specific:

    Keep SKILL.md concise and focused on activation, tasks, prompts, and guidance.

    line 138
  6. Distinctive:

    Avoid duplicating information across supporting documents.

    line 142
  7. Distinctive:

    Keep each document focused on a single responsibility.

    line 150
  8. Distinctive:

    Prefer reusable HR knowledge over organization-specific content.

    line 151
  9. Distinctive:

    Use descriptive headings instead of long paragraphs.

    line 152
  10. Distinctive:

    Keep prompt libraries easy to browse and maintain.

    line 154
  11. Distinctive:

    Use realistic HR scenarios in workflow examples.

    line 155
  12. Distinctive:

    Keep repository documentation synchronized whenever a skill changes.

    line 156

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill 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
SKILL.md

HR Skills repository guide

This skill defines the repository-wide conventions for the hr-skills monorepo. It helps AI assistants and contributors consistently create, validate, edit, and maintain skill packages.

Each skill is stored in its own directory under skills/ and must contain a SKILL.md. Supporting documentation may be added through content/, prompts/, and examples/, each serving a distinct purpose.

Repository structure

Each skill lives under skills/hr-<name>/.

skills/hr-<name>/
├── SKILL.md                    # Required
├── content/                    # Optional
│   └── <name>.md
├── prompts/                    # Optional
│   └── <topic>.md
└── examples/                   # Optional
    └── <scenario>.md

File responsibilities

Each file has a single responsibility.

  • SKILL.md defines activation behavior, supported tasks, prompt summaries, repository metadata, and usage guidance.
  • content/ contains detailed reference documentation, concepts, frameworks, terminology, and best practices.
  • prompts/ stores reusable prompt libraries grouped by a single HR topic.
  • examples/ demonstrates realistic end-to-end HR workflows and business scenarios.

Supporting files should complement each other instead of repeating the same information.

Content quality standards

  • HR-domain only — prompts and guidance must cover HR-specific patterns and best practices, not generic management or business advice.
  • No obvious content — avoid widely-known basics. Focus on nuanced guidance that HR professionals actually need AI assistance with.
  • SKILL.md is for AI agents — be concise; context window is a shared resource. Assume the agent is smart; only include what it doesn't already know. Use progressive disclosure — SKILL.md is an overview, not an exhaustive manual.
  • Trigger phrases matter — the description field determines when a skill activates. Include specific, realistic HR trigger phrases like "Write a PIP", "Conduct a stay interview", "Analyze turnover".

Checklist before publishing a skill

Core quality

  • description is specific and includes key HR trigger phrases
  • description covers both what the skill does and when to use it
  • SKILL.md body is under 500 lines
  • Prompts are grouped by meaningful subtopics
  • No time-sensitive information (laws, tools, versions)
  • Consistent HR terminology throughout
  • Prompts use [placeholders] for variable inputs
  • Each subtopic has 4-7 focused prompts
  • Tips section provides actionable professional guidance

Frontmatter

  • name matches the skill directory name exactly
  • description is wrapped in a single pair of double quotes, with trigger phrases as plain comma-separated text (not individually quoted)
  • metadata.author is set to Tuan Duc Tran
  • metadata.version is set

Structure

  • ## Supported tasks lists 8-12 concrete tasks
  • ## Key prompts is divided into logical subtopic sections
  • ## Tips has 4-6 professional best-practice tips

Supported tasks

  • Explain repository conventions, workflows, and directory structure
  • Validate skill packages against repository standards
  • Scaffold new skills following repository conventions
  • Review and improve SKILL.md
  • Review and improve content/, prompts/, and examples/
  • Validate frontmatter and naming conventions
  • Verify directory structure and file organization
  • Detect duplicated content across files
  • Validate Markdown formatting and repository rules
  • Verify internal relative links
  • Generate contributor guidance
  • Produce release and pre-publish checklists
  • Generate Conventional Commit messages
  • Recommend CI commands using Bun and Turbo
  • Keep repository documentation synchronized with repository changes

Key prompts

Repository management

  • "Explain the repository structure and file responsibilities."
  • "Describe the workflow for creating a new HR skill."
  • "Summarize the repository conventions contributors should follow."
  • "List the validation commands required before opening a pull request."
  • "Explain the differences between SKILL.md, content/, prompts/, and examples/."

Skill validation

  • "Review this skill package against repository standards."
  • "Validate this SKILL.md and list every issue."
  • "Check whether this skill follows naming conventions."
  • "Identify duplicated content across all supporting documents."
  • "Validate Markdown formatting and internal links."

Skill scaffolding

  • "Scaffold a complete HR skill package."
  • "Generate a new SKILL.md following repository conventions."
  • "Create a matching content/, prompts/, and examples/ structure."
  • "Generate missing supporting documents for an existing skill."
  • "Create repository-compliant frontmatter."

Content and prompt management

  • "Review this reference document for completeness."
  • "Generate a reusable prompt library for this HR topic."
  • "Organize these prompts into logical categories."
  • "Remove duplicated prompts already summarized in SKILL.md."
  • "Recommend where content should belong within the skill package."

Contributor assistance

  • "Generate a Conventional Commit message."
  • "Create a pull request checklist."
  • "Generate a GitHub Actions workflow for repository validation."
  • "Explain how to add a new skill from start to finish."
  • "Review this pull request for repository consistency."

Tips

  • Always include a SKILL.md in every skill package.
  • Keep SKILL.md concise and focused on activation, tasks, prompts, and guidance.
  • Store conceptual documentation in content/.
  • Store reusable prompt collections in prompts/.
  • Store realistic business workflows in examples/.
  • Avoid duplicating information across supporting documents.
  • Name directories and files using kebab-case.
  • Files inside content/ should omit the hr- prefix (for example, skills/hr-kpi/content/kpi.md).
  • Organize prompt libraries around a single HR topic.
  • Validate Markdown formatting, naming conventions, and internal links before committing.

Best practices

  • Keep each document focused on a single responsibility.
  • Prefer reusable HR knowledge over organization-specific content.
  • Use descriptive headings instead of long paragraphs.
  • Cross-reference related documents rather than duplicating information.
  • Keep prompt libraries easy to browse and maintain.
  • Use realistic HR scenarios in workflow examples.
  • Keep repository documentation synchronized whenever a skill changes.

Common mistakes

  • Putting long conceptual documentation inside SKILL.md.
  • Duplicating prompts between SKILL.md and prompts/.
  • Mixing multiple unrelated topics in one prompt library.
  • Using inconsistent skill or directory names.
  • Naming content/ files with the hr- prefix.
  • Forgetting to update internal links after renaming files.
  • Creating examples without realistic business context.
  • Allowing supporting documents to become inconsistent with SKILL.md.

Naming conventions

ItemConvention
Skill directoryskills/hr-kebab-case/
name fieldMatches the directory name
Content fileSkill name without the hr- prefix
Prompt fileDescriptive kebab-case topic
Example fileScenario-based kebab-case filename
Markdown headingsFollow sequential heading levels
Internal linksAlways use relative paths

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.