Total 30,812 skills, AI & Machine Learning has 4972 skills
Showing 12 of 4972 skills
Build, scaffold, refactor, and troubleshoot ChatGPT Apps SDK applications that combine an MCP server and widget UI. Use when Codex needs to design tools, register UI resources, wire the MCP Apps bridge or ChatGPT compatibility APIs, apply Apps SDK metadata or CSP or domain settings, or produce a docs-aligned project scaffold. Prefer a docs-first workflow by invoking the openai-docs skill or OpenAI developer docs MCP tools before generating code.
Use when auditing, trimming, or restructuring AI skill files to reduce context-window consumption. Trigger whenever a SKILL.md exceeds 120 lines, skills share duplicated content, AGENTS.md has large inline blocks, or the user asks to optimize, slim down, or reduce token usage of their skills.
Convert PRDs to prd.json format for the Ralph autonomous agent system. Use when you have an existing PRD and need to convert it to Ralph's JSON format. Triggers on: convert this prd, turn this into ralph format, create prd.json from this, ralph json.
General-purpose deep research with multi-source synthesis and confidence-scored findings. Auto-classifies complexity from quick lookup to exhaustive investigation. Cross-validates across independent sources with anti-hallucination verification, contradiction detection, and bias auditing. Produces synthesis products with evidence chains and provenance. Resumable journal sessions. Use when investigating technical topics, academic questions, market analysis, competitive intelligence, architecture decisions, technology evaluation, fact-checking, literature review, or trend analysis. NOT for code review (use honest-review), strategic decisions (use wargame), multi-perspective debate (use host-panel), or simple factual Q&A answerable in one search.
Agent definition conventions. Use when creating or modifying agents at any level (~/.claude/agents/, .claude/agents/, or project-local). Validate frontmatter, update README.md index. NOT for creating skills, MCP servers, or modifying CLAUDE.md.
Comprehensive prompt and context engineering for any AI system. Four modes: (1) Craft new prompts from scratch, (2) Analyze existing prompts with diagnostic scoring and optional improvement, (3) Convert prompts between model families (Claude/GPT/Gemini/Llama), (4) Evaluate prompts with test suites and rubrics. Adapts all recommendations to model class (instruction-following vs reasoning). Validates findings against current documentation. Use for system prompts, agent prompts, RAG pipelines, tool definitions, or any LLM context design. NOT for running prompts, generating content, or building agents.
Capture corrections, insights, and patterns as reusable project knowledge. Routes learnings to the right instruction file. Applies kaizen: small improvements, error-proofing, standards work. Auto-invoked when a correction pattern is detected 3+ times. Also use manually when Claude makes a repeated mistake, discovers a non-obvious gotcha, or when you want to persist a workflow preference.
RFC-driven multi-agent DAG execution pattern with quality gates, merge queues, and work unit orchestration.
通过逆向工程的 Gemini Web API 生成图片和文本。支持文本生成、提示词生图、参考图片视觉输入和多轮对话。当其他技能需要图片生成后端，或用户要求"用Gemini生成图片"、"Gemini文本生成"时使用。
Operate long-lived agent workloads with observability, security boundaries, and lifecycle management.
通过兔子API（nano-banana 模型）、Google、OpenAI、DashScope 和 Replicate 进行 AI 图片生成。支持文生图、参考图片、宽高比、模型选择。当用户要求生成、创建或绘制图片时使用。
Transcribe audio files using Groq API (Whisper models). Use when user needs to transcribe audio to text.