Skills for AI coding agents.
make install| Agent | Skills Path |
|---|---|
| Claude Code | ~/.claude/skills/ |
| OpenCode, Pi, Codex | ~/.agents/skills/ |
These methodology skills improve AI agent effectiveness:
| Skill | Description |
|---|---|
brainstorming |
Explore ideas before implementation through collaborative dialogue |
writing-plans |
Create detailed implementation plans with bite-sized tasks |
executing-plans |
Execute plans task-by-task with verification |
test-driven-development |
Write tests first, watch them fail, implement minimally |
systematic-debugging |
Find root cause before attempting fixes |
verification-before-completion |
Evidence before claims, always |
dispatching-parallel-agents |
Run multiple independent investigations concurrently |
receiving-code-review |
Technical rigor when implementing feedback |
requesting-code-review |
Verify work meets requirements before merging |
semantic-commit |
Conventional commits for clear history |
fetch-ci-build |
Fetch CI build results and diagnose failures across CI providers |
| Command | Description |
|---|---|
make install |
Build and install skills for all agents |
make build |
Build skills to build/ without installing |
make install-skills |
Install skills only |
make clean |
Remove all installed artifacts |
make help |
Show all available commands |
- Create
skills/<skill-name>/SKILL.mdwith YAML frontmatter - Add supporting files to the same directory
- Run
make install
See Agent Skills specification for format details.
- Python 3.11+