Production-grade Claude Code configuration: 23 agents, 27 skills, 12 safety hooks, MCP integrations. A complete guide to building an enterprise-level Claude Code setup.
Runs a structured multi-round interview to gather the user's preferences, context, and decisions on a topic — researching before and between rounds, opening an HTML briefing in cmux before each round, and closing with a summarize-back and a deliverable (spec, goal.md, plan, or decision log). Use when the user says…
Stage, commit, and push changes following project conventions. Stages specific files (never git add . or -A), writes conventional commit messages, and pushes to remote.
Analyzes recent Claude Code session transcripts and project activity to identify recurring patterns, then recommends new skills, agents, or workflow improvements based on observed behavior. Example usage.
Multi-agent orchestration skill that spawns parallel specialist agents to analyze a feature from architecture, testing, and security perspectives, then synthesizes a unified implementation plan. Example usage.
Trigger the 9-person Executive Reviewer Panel to evaluate the product roadmap. Runs all reviewers in parallel, synthesizes a scored dashboard, and presents recommendations for you to accept, reject, or defer. Example usage.
★not rated 2 2mo agoA97 tokens
originalMIT
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: