Agent skill authoring

Reusable AI agent skills, project templates, and guardrails for safer software maintenance and delivery.

Install

npx -y skills add jukrap/ai-agent-playbook --skill agent-skill-authoring

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

Copied from the file, not written here

Use when creating, reviewing, or reorganizing reusable agent skills, skill references, trigger descriptions, or skill/template boundaries.

1.9 KB, as published. Nobody here has run it

Create small reusable skills that are easy for agents to discover and hard to misuse.

Confirm the behavior is reusable across projects. If it is project-specific standing policy, put it in a template or project doc instead.
Write the trigger first: description starts with Use when... and describes when to load the skill, not the workflow.
Keep SKILL.md concise: trigger, core workflow, and when to read references.
Move long rules, examples, API details, and checklists into one-level references/*.md files.
Add scripts only for deterministic repeated work such as validation, formatting, or artifact generation.
Test the trigger with realistic requests and near misses; tighten wording if the skill would load too often.
Check whether an existing skill, template, or project guide can be strengthened before adding another skill.
Update indexes, translations, validation metadata, and installed copies through the repository maintenance workflow.

Do not turn every good rule into a skill.
Do not put product scope, milestone status, branch policy, personal paths, or machine-specific setup into installable skills.
Do not write a long skill that duplicates project AGENTS.md, local docs, or an existing process skill pack.
Do not vendor an external skill pack wholesale when a short local rule or renamed focused skill would be clearer.
Do not rely on one agent runtime's hooks, slash commands, or environment variables in agent-agnostic skill source.

Read references/skill-shape-and-routing.md before creating or reorganizing a skill.

Read references/reference-authoring-checks.md before adding long references, examples, stack details, or adopted external patterns.