Loading...
Loading...
Found 6,669 Skills
Use when asked to detect silent failures/weak error handling or explicitly asked to run the silent-failure-hunter subagent.
Use when executing tasks from a task breakdown document in the current session.
Comprehensive technical research by combining multiple intelligence sources — Grok (X/Twitter developer discussions via Playwright), DeepWiki (AI-powered GitHub repository analysis), and WebSearch. Dispatches parallel subagents for each source and synthesizes findings into a unified report. This skill should be used when evaluating technologies, comparing libraries/frameworks, researching GitHub repos, gauging developer sentiment, or investigating technical architecture decisions. Trigger phrases include "tech research", "research this technology", "技术调研", "调研一下", "compare libraries", "evaluate framework", "investigate repo".
Use when answering complex questions about a codebase that require exploring multiple areas or understanding how components connect - coordinates parallel sub-agents to locate, analyze, and synthesize findings
Fetches the latest news using news-aggregator-skill, formats it into a podcast script in Markdown format, and uses the tts skill to generate a podcast audio file. Use when the user asks to get the latest news and read it out as a podcast.
Audit Claude Code configuration health across all layers (CLAUDE.md, rules, skills, hooks, MCP). Run periodically or when collaboration feels off.
Use git-agent to commit changes with AI-generated conventional commit messages. Immediately runs git-agent commit when loaded — no setup or configuration questions unless an error occurs.
Post-task review. Extract learnings, classify, write to memory layers, and reconcile GitHub issues.
Turn your AI into a full design team — 17 specialists that research, strategize, write, design, build, and review.
Design multi-agent harnesses for long-running autonomous coding tasks. Covers generator/evaluator loops, context reset strategy, sprint contracts, and the planner-generator-evaluator architecture from Anthropic's harness research.
Run /check before merge to confirm a change is sound. Two modes: `/check verify` drives the real app and proves behavior against the spec (every acceptance criterion met, every specced surface built); `/check review` runs a senior code review on a different model than wrote the code. Verify after /develop, review before a PR. Writes findings to docs/reviews/; never edits your code.
Build resumable multi-agent workflows with durable execution, tool loops, and automatic stream recovery on client reconnection.