oz-skills
AGENTS.md825 starsChanged 7 months agoMITView on GitHub
This file provides guidance to WARP (warp.dev) when working with code in this repository. This is the oz-skills repository - a public catalog of reusable Agent Skills for Warp AI agents and Oz. These are pre-built skills that external users can copy into their projects to teach agents about common workflows and best practices. This repository follows the Agent Skills open standard. Agent Skills are folders containing a SKILL.md file with YAML frontmatter and markdown instructions that agents can discover…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository15 other files it gives its agents.
- analysis-artifactsSkill
- ci-fixSkill
- create-pull-requestSkill
- dbt-model-indexSkill
- docs-updateSkill
- github-bug-report-triageSkill
- github-issue-dedupeSkill
- mcp-builderSkill
- schedulerSkill
- seo-aeo-auditSkill
- slack-qa-investigateSkill
- terraform-style-checkSkill
- web-accessibility-auditSkill
- webapp-testingSkill
- web-performance-auditSkill
The file
We split this AGENTS.md into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
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
- Very specific: line 17
All skills are stored in the .agents/skills/ directory.
- Distinctive: line 52
Use these rules whenever creating or editing a skill:
- Very specific: line 58
Keep description concise and concrete.
- Distinctive: line 64
Follow this template when authoring a new skill:
- Distinctive: line 95
When working in this repository:
- Distinctive: line 110
Verify the agent follows the skill's guidance
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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.
Files
AGENTS.md
This file provides guidance to WARP (warp.dev) when working with code in this repository.
Repository Overview
This is the oz-skills repository - a public catalog of reusable Agent Skills for Warp AI agents and Oz. These are pre-built skills that external users can copy into their projects to teach agents about common workflows and best practices.
Agent Skills Standard
This repository follows the Agent Skills open standard. Agent Skills are folders containing a SKILL.md file with YAML frontmatter and markdown instructions that agents can discover and use.
For comprehensive details about the skills format and usage patterns, see wip/warp-skills-overview.md.
Repository Structure
All skills are stored in the .agents/skills/ directory. Each skill has its own subdirectory:
.agents/skills/
├── docs-update/
│ └── SKILL.md
├── another-skill/
│ └── SKILL.md
└── ...Skill File Format
Each skill must have a SKILL.md file with this structure:
---
name: skill-name
description: Imperative summary sentence. Use when sentence.
---
# Skill Title
Your instructions, examples, and guidance here in markdown format.Required Fields
- name: Kebab-case identifier (e.g.,
docs-update,rust-testing,api-conventions) - description: Two sentences in a strict pattern:
- Sentence 1: Imperative verb + what the skill does (e.g.,
Audit...,Build...,Test...) - Sentence 2: Starts with
Use when...and describes trigger scenarios
- Sentence 1: Imperative verb + what the skill does (e.g.,
Skill Metadata Style Guide (Canonical)
Use these rules whenever creating or editing a skill:
namemust be kebab-case and exactly match the skill directory namedescriptionmust be exactly two sentencesdescriptionsentence 1 must start with an imperative verb (Build,Audit,Test,Optimize,Triage,Generate,Investigate, etc.)descriptionsentence 2 must start withUse when...- Keep
descriptionconcise and concrete; avoid filler and unnecessary quotes SKILL.mdmust include a single H1 in Title Case- H1 should reflect the skill's human-readable purpose (for example,
# Web Performance Audit)
Authoring Template (Copy/Paste)
Follow this template when authoring a new skill:
---
name: your-skill-name
description: Verb-first summary of what this skill does. Use when the user asks for the workflow this skill is designed to handle.
---
# Your Skill Title
## When to Use
- Trigger condition 1
- Trigger condition 2
- Trigger condition 3
## Workflow
1. Step one
2. Step two
3. Step three
## Examples
```bash
# Example command(s)
```Repository Conventions
When working in this repository:
- One skill per directory in
.agents/skills/ - Use kebab-case for skill directory names (e.g.,
docs-update,deployment-workflow) - Each skill must have a
SKILL.mdfile with proper frontmatter - Keep descriptions concise - they help agents decide when to load the skill
- Include examples - real code snippets and commands make skills more useful
- Focus on workflows - skills should capture procedural knowledge, not general documentation
Testing Skills
To test a skill from this repository:
- Copy the skill folder to a test project's
.agents/skills/directory - Interact with a Warp agent on a relevant task
- Verify the agent follows the skill's guidance
- Refine the skill based on results
Common Skill Types
Good candidates for skills in this repository:
- Workflow automation (e.g., docs-update for automatic documentation updates)
- Testing patterns (e.g., how to write tests for specific frameworks)
- Deployment procedures (e.g., CI/CD workflows)
- Code conventions (e.g., style guides, naming patterns)
- Tool usage (e.g., how to use specific CLIs or APIs)
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
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.

