发现 Skills

Comprehensive academic textbook chapter writing system for medical and scientific content. Use when the user wants to: (1) write a full textbook chapter (5,000–15,000 words) on any medical or scientific topic, (2) generate a detailed table of contents with per-section word counts, (3) research topics via PubMed MCP and compile 20–30 references, (4) write section-by-section with proper citations in Vancouver format, (5) create publishable academic content in an Eric Topol–inspired voice with authentic human prose, (6) get approval at the TOC stage before writing begins, and (7) export well-structured chapters ready for textbook publication.

Use OpenAI Codex CLI for coding tasks. Triggers: codex, code review, fix CI, refactor code, implement feature, coding agent, gpt-5-codex. Enables Clawdbot to delegate coding work to Codex CLI as a subagent or direct tool.

Process large datasets efficiently using chunk(), chunkById(), lazy(), and cursor() to reduce memory consumption and improve performance

Generative Engine Optimization for AI search engines (ChatGPT, Claude, Perplexity).

Create and manage sprite animations with multiple frames, animation tags, frame durations, and linked cels. Use when the user wants to animate a sprite, add animation, create movement, make it move, mentions "animation", "animated", "frames", "keyframes", "frame rate", "FPS", "timing", "duration", "walk cycle", "run cycle", "idle animation", "attack animation", "jump", "movement", "motion", or describes actions like "walking", "running", "jumping", "attacking", "breathing", "bobbing", "bouncing". Trigger on animation tags, loops, playback, sequences, "add frames", "duplicate frame", "frame timing", "ping-pong", "loop", "sequence". Also for linked cels, static backgrounds, and frame optimization.

Query Polymarket prediction market odds and events via CLI. Search for markets, get current prices, list events by category. Supports sports betting (NFL, NBA, soccer/EPL, Champions League), politics, crypto, elections, geopolitics. Real money markets = more accurate than polls. No API key required. Use when asked about odds, probabilities, predictions, or 'what are the chances of X'.

This skill should be used when the user asks to "diagnose context problems", "fix lost-in-middle issues", "debug agent failures", "understand context poisoning", or mentions context degradation, attention patterns, context clash, context confusion, or agent performance degradation. Provides patterns for recognizing and mitigating context failures.

Generate and edit images using the Gemini API (Nano Banana). Use this skill when creating images from text prompts, editing existing images, applying style transfers, generating logos with text, creating stickers, product mockups, or any image generation/manipulation task. Supports text-to-image, image editing, multi-turn refinement, and composition from multiple reference images.

Use when deciding whether a service is safe to launch, scale, or recover in production. USE FOR: run production readiness review, check rollback and canary plan, assess disaster recovery readiness, review incident response runbooks, evaluate release go-live checklist. DO NOT USE FOR: day-to-day bug fixing, early product ideation without deployment scope.

See monitoring.md for general monitoring principles (the three pillars of observability, metric types/naming, key metrics, alerting rules, health-check concepts, and anti-patterns).

Review the code of a feature or a file in a rigorous and practical manner. Identify issues by severity and propose improvements, without applying changes without authorization.

Deploy Caddy reverse proxy on remote servers with automatic SSL and systemd integration. Use when users want to: (1) set up a reverse proxy for local web services, (2) configure automatic Let's Encrypt SSL certificates, (3) create a systemd service with auto-start, and (4) proxy HTTP and WebSocket traffic. Triggered by phrases like "deploy caddy", "setup reverse proxy", "configure caddy", "caddy ssl".

A pure Markdown directive skill used to design and create Codex multi-agents required for a project. Using only SKILL.md and reference materials, it carries out project analysis → specification/case verification → design of agent_plan.tsv → manual guidance for applying changes to .codex → verification reporting.

Design and optimize AI agent action spaces, tool definitions, and observation formatting for higher completion rates.

A comprehensive verification system for Claude Code sessions.

Security anti-pattern for verbose error messages (CWE-209). Use when generating or reviewing code that handles errors, exceptions, or generates user-facing error responses. Detects stack trace exposure and detailed error information leakage to users.

Coordination protocol for the main Claude Code agent. Explicit user invocation required ("mobilize agents", "coordinate", "check registry"). Provides agent orchestration, registry management, and handoff protocols. Subagents never access this — the main agent supplies context in task prompts.

ZenTao Product (Product) management — list products and retrieve product details. A product is the top-level entity in ZenTao that contains requirements and bugs. Use this when a user needs to view the product list, see the details of a specific product, or filter requirements under a particular product.

Slack integration for team communication. Use when searching for context in channels, reading channel history, posting messages, or finding discussions about topics.

Optimize AI engine visibility. Use when: AEO/GEO strategy, citation optimization, entity consistency across AI platforms.

LinkedIn post writing with hook formulas, formatting rules, and engagement patterns. Covers post types, algorithm signals, character limits, and content pillars. Use for: LinkedIn posts, professional content, thought leadership, B2B content, personal branding. Triggers: linkedin post, linkedin content, linkedin writing, linkedin strategy, linkedin engagement, linkedin algorithm, linkedin hook, linkedin formatting, thought leadership, professional content, b2b content, linkedin growth

Refine texts in any language: perfect grammar and spelling, paraphrase ideas, avoid AI detection while maintaining an authentic voice. Detects grammatical patterns that signal cognitive states or structural issues—acting as a sensory system for deeper strategic insights.

Use this skill for web search, extraction, mapping, crawling, and research via Tavily’s REST API when web searches are needed and no built-in tool is available, or when Tavily’s LLM-friendly format is beneficial.

Post entries and search the shared Hermes notebook