ai

Planner-Generator-Evaluator 编排模式。GAN 启发的多 agent 动态编排，通过 ai CLI 控制子 agent 完成复杂任务的拆解-执行-验证闭环。

Code review skill using codex-rs methodology with ai CLI

为任意代码仓库构建结构化知识库（code wiki），并通过 frozen snapshot 机制启动即问即答的 expert agent。 当你需要深入理解一个大型代码仓库、构建可复用的代码知识库、或为其他 agent 提供代码专家服务时使用此技能。

Use the mcporter CLI to list, configure, auth, and call MCP servers/tools directly (HTTP or stdio), including ad-hoc servers, config edits, and CLI/type generation.

Orchestrate a real-time alternating debate between two subagents using ai serve/send. Judge controls rounds by prompting each agent in turn.

任务规划与进度追踪。当任务有 5+ 步骤、多文件改动、或长执行链时使用。 核心机制：用 checkbox 列表追踪进度，每完成一步立即更新，利用"重写列表"将完整任务清单重新注入 context。

使用 ai serve/send/watch/kill 控制子 agent 的通用指引。所有需要子 agent 的技能都应参考此技能，而非重复定义 spawn/cleanup 流程。

Generate a rich search index of all skills at ~/.ai/skill-index.json. Use when skills change or when /skills reindex is triggered. Produces aliases, use-when triggers, and categories for each skill via LLM intelligence. Supports incremental updates.

Explore codebases, repositories, or topics and collect key information for later phases. Use for code exploration, architecture analysis, and information gathering before implementation.

Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification

Worker-Judge 迭代循环。一个 agent 产出，另一个独立 agent 审查，循环直到通过。适用于需要质量保证的任何任务（规划、实现、审查等）。

通过对话探索用户意图，产出 design.md。design 覆盖 5 个内容维度，让无上下文的 subagent 也能理解全貌。

Use tmux for background tasks, TUI testing, and debugging layouts. Philosophy: No background bash - use tmux for full observability and direct interaction.

Fast code search using Semble. Use when you need to understand a codebase, find implementations, locate symbols, or explore unfamiliar code. Replaces grep+read for exploratory searches with ~98% fewer tokens. Works fully offline after first model download.

Fetch and read transcripts from YouTube and Bilibili videos. Use when you need to summarize a video, answer questions about its content, or extract information from it.

Clean up ~/.ai/ disk space (runs, sessions, traces). Use when user says "清理", "磁盘空间", "cleanup", or wants to free space under ~/.ai/.

Self-governance protocol for autonomous agents: WAL (Write-Ahead Log), VBR (Verify Before Reporting), ADL (Anti-Divergence Limit), VFM (Value-For-Money), and IKL (Infrastructure Knowledge Logging). Use when: (1) receiving a user correction — log it before responding, (2) making an important decision or analysis — log it before continuing, (3) pre-compaction memory flush — flush the working buffer to WAL, (4) session start — replay unapplied WAL entries to restore lost context, (5) any time you want to ensure something survives compaction, (6) before claiming a task is done — verify it, (7) periodic self-check — am I drifting from my persona? (8) cost tracking — was that expensive operation worth it? (9) discovering infrastructure — log hardware/service specs immediately.

This skill should be used when users need to scrape websites, extract structured data, handle JavaScript-heavy pages, crawl multiple URLs, or build automated web data pipelines. Includes optimized extraction patterns with schema generation for efficient, LLM-free extraction.

搜索和安装技能。当用户问"有没有 X 的 skill"、"搜索 skill"、"安装 skill"时使用。

Query latest API documentation to avoid LLM hallucinations. Context7 provides real-time access to the latest library and framework documentation, ensuring AI-generated code uses current APIs rather than outdated training data.

Create data-driven presentation slides using React, Vite, and Recharts with Sentry branding. Use when asked to "create a presentation", "build slides", "make a deck", "create a data presentation", "build a Sentry presentation". Scaffolds a complete slide-based app with charts, animations, and single-file HTML output.

Reverse engineer desktop applications, binaries, and software packages. Use when the user asks to analyze, reverse engineer, or understand how an application works internally. Targets include macOS .app bundles, Electron apps, Tauri apps, native binaries, npm/pip packages, CLI tools, and any software the user wants to understand the architecture and implementation of.

EvoClaw Tiered Memory Architecture v2.2.0 - LLM-powered three-tier memory system with automatic daily note ingestion, structured metadata extraction, URL preservation, validation, and cloud-first sync.

Maintain and query a personal LLM Wiki — an incremental, compiled knowledge base of structured Markdown files.

Merge a PR automatically with safety checks.

Find deepening opportunities in a codebase. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup

Interact with Symphony task orchestration system. Create tasks, check status, manage workflow automation. Use for AI-driven bug fixes, features, and automation.

Use during IMPLEMENTATION phase. Write test first, watch it fail, then write minimal code to pass. This is HOW you write code, not WHEN to start a project.

Write system prompts, tool docs, and agent definitions. Combines research-backed prompt engineering (+15-30% measured improvements) with project XML conventions. Covers tag hierarchy, structural templates, high-impact interventions, anti-patterns.

Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.