Comprehensive configuration system for Claude Code
This repository also configures its own agents. See what claude-code tells them →
Command Claude Code
Use when the user signals they're stepping away (brb, afk, heading out) — keep working autonomously, skip anything that needs interactive approval, handoff when done or blocked.
Command Claude Code
Review the changes in the repo and prepare logical, organized, clean and descriptive git commits using conventional commits.
Command Claude Code
Find and fix issue $ARGUMENTS.
Command Claude Code
Sequenced tilt bootstrap then e2e test baseline improvement. Runs tiltup workflow, then e2e test loop with categorized fixes and conventional commits. Not for unit-test-only projects or non-tilt dev environments.
Command Claude Code
Generate concise handoff summary with context.
Command Claude Code
Reimplement the current branch with a clean, narrative-quality git commit history suitable for reviewer comprehension. Execute each step fully before proceeding to the next.
Command Claude Code
Interview to fill gaps in a specification file.
Command Claude Code
Interactive Linear issue triage with codebase verification. Lists issues, verifies implementation status via Explore agents, and updates issue states.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: