AlexAnys

Use when distilling a bounded framework capability — a subsystem with state machine, external protocol (CLI/API/socket/process), multiple adapters, async/streaming, cancellation/timeout, or provider drift risk. Builds a fake harness BEFORE any local implementation, then verifies via conformance tests. Do NOT use for narrow pure functions; use function-distiller for those.

Research an external or internal software repo, distill one narrow module/function into a minimal self-contained implementation, and verify equivalence without copying code. Use for dependency removal, framework research, API-compatible local reimplementation, DeepWiki/GitHub-based source study, or clean-room behavior specs of pure or near-pure functions. For bounded framework capabilities (state machine + protocol + adapters + async), use capability-distiller instead.

Use when distilling a bounded framework capability — a subsystem with state machine, external protocol (CLI/API/socket/process), multiple adapters, async/streaming, cancellation/timeout, or provider drift risk. Builds a fake harness BEFORE any local implementation, then verifies via conformance tests. Do NOT use for narrow pure functions; use function-distiller for those.

Research an external or internal software repo, distill one narrow module/function into a minimal self-contained implementation, and verify equivalence without copying code. Use for dependency removal, framework research, API-compatible local reimplementation, DeepWiki/GitHub-based source study, or clean-room behavior specs of pure or near-pure functions. For bounded framework capabilities (state machine + protocol + adapters + async), use capability-distiller instead.

Connect a Feishu (Lark) bot to Clawdbot via WebSocket long-connection. No public server, domain, or ngrok required. Use when setting up Feishu/Lark as a messaging channel, troubleshooting the Feishu bridge, or managing the bridge service (start/stop/logs). Covers bot creation on Feishu Open Platform, credential setup, bridge startup, macOS launchd auto-restart, and group chat behavior tuning.

飞书全能 MCP - 创建/编辑文档、插入图表、管理表格、搜索知识库。整合社区版+官方 MCP，适配 Moltbot/Clawdbot。

Design and set up a multi-agent harness for any project type — software development (Planner → Builder → QA), knowledge compilation (Coordinator → Compiler → QA), or operations (Coordinator → Executor → Monitor). Use this skill whenever the user wants to build with multiple AI agents, set up automated workflows, create an agent pipeline, apply Anthropic's harness methodology, or structure any complex task that benefits from separating planning, execution, and verification. Covers software projects, knowledge bases/wikis, research systems, operational pipelines, and hybrids. Also use when the user mentions "harness", "multi-agent", "planner/builder/qa", "coordinator/compiler", "generator/evaluator", or wants to coordinate multiple Claude Code instances.

Design and set up an agent harness for any project — first judge the verification signal (can success be machine-checked? are failures reversible? can the work be split?), then choose a structure: solo agent + check command, pipeline, Planner → Builder → QA triangle, or fan-out workers. The triangle is one structure, not the default. Use this skill whenever the user wants to build with multiple AI agents, set up automated workflows, create an agent pipeline, build an eval harness or an experiment/optimization loop, apply harness methodology, or structure any complex task that benefits from separating planning, execution, and verification. Also use when the user mentions "harness", "multi-agent", "planner/builder/qa", "coordinator/compiler", "generator/evaluator", or wants to coordinate multiple Claude Code instances.