Find the perfect skill
Commit message generation with gitmoji
Development
Generates commit messages from staged changes using the repository's commit convention, including gitmoji, type, scope, and optional actual commit.
Mirra Voice
Development
Voice call history, transcripts, and call metadata via REST API.
Skill Creator for Claude Code
Development
Guide for creating effective skills that extend Claude's capabilities with specialized knowledge, workflows, and tool integrations.
Explain code with diagrams and analogies
Development
Explains code using diagrams and analogies to make understanding easier and more intuitive.
PR Lifecycle Workflow
Development
Drive the full PR lifecycle: create or reuse a PR, request Copilot review, monitor CI checks, and summarize findings for the user.
Code Review Orchestrator
Development
Orchestrates specialized subagents to perform a comprehensive code review and produce a markdown report. Works in PR mode (with --pr) or on the current branch, analyzing changes and invoking reviewers for Go, architecture, conventions, Bubble Tea, and documentation.
Automated codebase exploration
Development
Automatically launches parallel exploration agents to analyze the codebase: similar features, architecture, integrations. Consolidates findings into a workflow state with no user interaction required.
Multi-persona code reviewer
Development
Multi-language code reviewer with 9 specialized personas, works locally or on GitHub PRs, posts inline comments and resolves its own prior comments.
Create a reusable skill
Development
Create a reusable skill to automate repetitive tasks in Catalyst.
Sentry Common Pitfalls
Development
Execute common Sentry pitfalls and how to avoid them. Use when troubleshooting Sentry issues, reviewing configurations, or preventing common mistakes.
Import Figma to Lumos
Development
Translate a Figma design into Lumos tokens for Astro, handling conversions and reconciling inconsistencies without inventing variables.
ADR Review Level 1 (Runcriticon)
Development
Apply the ultrathink Level 1 review pattern to an ADR from the Runcriticon corpus: detect gaps, propose sub-decisions, ask questions, and generate review and acceptance PRs.