| name | author-command |
| description | Authors a brand new command for the agent framework following project best practices and conventions. Trigger on user queries such as: 'Create a new command to <do something>', 'Author a command that <does something>', 'I need a command that <does something>'. |
Author Command
Overview
Author a new command for the agent framework following project best practices and conventions.
When To Use
User requests a new command ("Create a command to ", "I need a command that...").
Behavior
- Review command authoring docs and examples
- Gather requirements: functionality, inputs, outputs, constraints
- Draft command in markdown following example format
- Review with user, revise until approved (do not proceed without approval)
- Place final command in
./command/ directory
- Document in README.md
- Verify the new command against the full project test suite via the test skill (
/skill:test — run → triage → evaluate → loop until green) before declaring completion, so the authored command does not break existing behaviour.
Framework placeholders
$ARGUMENTS — full argument string$1, $2... — positional arguments!command — inject command stdout into prompt (use sparingly)@path/to/file — include file contents in prompt
Scripts
No CLI runner script. Use Pi prompt invocation or agent command framework.
- Template:
./assets/command-template.md
Policy
- Prefer canonical scripts where available
- No ad-hoc commits/pushes without explicit approval
Examples
/skill:author-command "Create a command to format dates for display"
wl show SA-0MPYMFZXO0004ZU4 --json
End.
