◆AI Engineer pack

Use when authoring an Agent Skill SKILL.md: follow the agentskills.io spec — WHEN: keyword with quoted triggers, lowercase-hyphen name, 60-word description, references/ and scripts/ dirs.

Query expansion, HyDE, and multi-query generation for improved retrieval

Content moderation API integration using OpenAI Moderation, Perspective API, and others

LlamaIndex agent and query engine setup for RAG-powered agents

Use when the user wants to add a configurable AI chatbot/assistant to a web app - a help bubble in the corner, a support bot, or an in-game NPC dialog - with a custom persona, scope guardrails, and static knowledge, powered by Gipity's LLM service with no API keys.

Structured messaging protocols for agent team communication including message type selection, plan approval, shutdown procedures, and anti-patterns to avoid. Use this skill when establishing communication norms for a newly spawned team, when deciding whether to send a direct message or a broadcast, when a team-lead needs to review and approve an implementer's plan before work begins, when orchestrating a graceful team shutdown after all tasks are complete, or when debugging why teammates are not

Use this skill when writing, improving, or debugging prompts for any AI model — Claude, GPT, Gemini, or Copilot. Also use when designing plan-before-execute workflows, verification patterns, self-improvement loops, or structuring AI instructions for maximum clarity and output quality. Trigger on keywords: prompt, prompt engineering, instruct AI, system prompt, few-shot, chain-of-thought, AI instruction, prompt template, prompt pattern.

Generate structured AI agent skills (SKILL.md) and llms.txt from your TypeScript API documentation

Applies the reasoning of David Silver, lead researcher on AlphaGo and AlphaZero at DeepMind, to problems of AI design, reinforcement learning, and open-ended discovery. Use this skill whenever you are designing AI systems, evaluating learning algorithms, balancing exploration vs. exploitation, choosing research problems, or discussing how to break past human performance ceilings. Reach for this whenever the user asks about self-play, Monte-Carlo Tree Search, tabula rasa learning, AGI, or moving

A minimal example skill that greets the user. Use when you want a tiny, valid reference skill to test skref against or to learn the SKILL.md format.

This skill should be used when the user asks to "create an MCP App", "add a UI to an MCP tool", "build an interactive MCP View", "scaffold an MCP App", or needs guidance on MCP Apps SDK patterns, UI-resource registration, MCP App lifecycle, or host integration. Provides comprehensive guidance for building MCP Apps with interactive UIs.

Query available AI models across all configured providers. List models, resolve model aliases, and browse the full model catalog including provider-specific variants.

Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.

Create and manage routing combos with 14 strategies (priority, weighted, round-robin, Auto-combo, etc.). Configure fallback chains, test routing outcomes, and retrieve combo metrics.

Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.

Assists with storing, searching, and managing vector embeddings using ChromaDB. Use when building RAG pipelines, semantic search engines, or recommendation systems. Trigger words: chromadb, chroma, vector database, embeddings, semantic search, similarity search, vector store, rag.

Use when testing invalid context validation

Develop, install, troubleshoot, and integrate OpenCode plugins for the opencode CLI. Use when: writing new OpenCode plugins, porting Pi SDK extensions to OpenCode, registering custom tools via the plugin API, fixing plugin load failures, resolving module dependency issues in plugins, installing plugins globally or per-project, understanding the plugin hook system, or any task involving @opencode-ai/plugin, opencode plugin command, or opencode.json plugin config. Covers: plugin format, Hooks API,

Design and build AI agents for any domain. Use when users: (1) ask to "create an agent", "build an assistant", or "design an AI system" (2) want to understand agent architecture, agentic patterns, or autonomous AI (3) need help with capabilities, subagents, planning, or skill mechanisms (4) ask about Claude Code, Cursor, or similar agent internals (5) want to build agents for business, research, creative, or operational tasks Keywords: agent, assistant, autonomous, workflow, tool use, multi-step

Design, implement, and debug autonomous AI agents and multi-agent systems using the Google Antigravity (AGY) SDK. ACTIVATE this skill when the user wants to create, configure, or orchestrate Google Antigravity agents.

Generate a complete MCP server project in TypeScript with tools, resources, and proper configuration

Build high-performing OpenClaw agents end-to-end with comprehensive safety features. Use when you want to design a new agent (persona + operating rules) and generate required OpenClaw workspace files (SOUL.md, IDENTITY.md, AGENTS.md, USER.md, HEARTBEAT.md, optional MEMORY.md + memory/YYYY-MM-DD.md). Includes anti-deadlock protection, timeout handling, error recovery, loop breaker, message overload protection, token limit protection, retry mechanism, health check, degraded mode, monitoring & logg

PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation.

Play LunchTable-TCG, a Yu-Gi-Oh-inspired online trading card game with AI agents

Design, implement, and debug autonomous AI agents and multi-agent systems in Rust using the Google Antigravity (AGY) Rust SDK. ACTIVATE this skill when the user wants to create, configure, or orchestrate Google Antigravity agents in Rust.

Fact-forcing gate that blocks Edit/Write/Bash (including MultiEdit) and demands concrete investigation (importers, data schemas, user instruction) before allowing the action. Measurably improves output quality by +2.25 points vs ungated agents.

This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.

This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.

Use when auditing how an AI/ML personnel assessment is described and how it affects people — Components 7-9 (information & perceptions) of the Landers & Behrend (2023) framework. Covers first-party developer claims (do they honestly and transparently follow from the audit evidence?), second-party effects on those assessed (candidate reactions, justice, false positives vs. false negatives, what is communicated), and third-party understanding (employment-law experts, regulators, community, public)

Machine Learning Interpretability: apply the codified rules and ground precise facts at runtime instead of recalling them.

You are an expert in AG2 (formerly AutoGen), the open-source multi-agent conversation framework. You help developers build systems where multiple AI agents collaborate through structured conversations — with tool use, human-in-the-loop, code execution, group chat orchestration, and nested conversations — for complex tasks like software development, research, and data analysis.

This skill should be used when the user asks to "create a plugin", "scaffold a plugin", "understand plugin structure", "organize plugin components", "set up plugin.json", "use ${CLAUDE_PLUGIN_ROOT}", "add commands/agents/skills/hooks", "configure auto-discovery", or needs guidance on plugin directory layout, manifest configuration, component organization, file naming conventions, or Claude Code plugin architecture best practices.

Generate, edit, and extend videos with xAI's grok-imagine-video model. Use when the user wants to create a video from text, animate a still image, build a reference-driven video (virtual try-on, character consistency), edit an existing video, or extend one. Asynchronous — the script handles polling.

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support

Track ML experiments with Trackio using Python logging, alerts, and CLI metric retrieval.

List, auth, and call MCP servers/tools from the terminal.

Use when auditing the data and foundational design of an AI/ML personnel assessment — Components 1-2 of the Landers & Behrend (2023) framework. Covers input-data population, sampling, range restriction, and incumbent-vs-applicant generalizability; and model design: how the criterion ("ground truth") is defined and its construct validity, why each predictor/feature was included, and whether choices were theory-driven or empirically derived. Triggers: "audit training data", "is the training sample

Split documents into overlapping token chunks for RAG pipelines and LLM context windows. Zero dependencies.

You are an expert in BrowserBase, the cloud platform for running headless browsers at scale. You help developers deploy browser-based automations, AI agents, and web scraping pipelines using managed Chromium instances with residential proxies, session recording, stealth mode, and parallel execution — without managing browser infrastructure.

Use when creating or configuring Claude Code agents and their frontmatter.

Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.

Instrument an AI agent or LLM app with OpenTelemetry and watch its runs live in tracelet (local DevTools). Use when asked to add tracing/observability to an agent, or to debug what an agent actually did.

Comprehensive Cline SDK skill for building AI agents. Covers the Agent runtime, ClineCore sessions, custom tools, plugins, events, LLM providers, scheduling, multi-agent teams, and production deployment. Use for any task involving @cline/sdk or its sub-packages.

Implement Conversational Language Understanding (CLU) using the azure-ai-language-conversations Python SDK. Use when working with ConversationAnalysisClient to analyze conversation intent and entities, building NLP features, or integrating language understanding into applications.

Referee for writer/reviewer loops between coding agents, where one tool writes code and the other reviews it in rounds: Claude Code writing while Codex reviews, or the reverse. Use when running or setting up a cross-model review loop, a test-fix loop, or any "loop until clean" workflow. Sets the loop contract up front, keeps an append-only per-round ledger, applies stop rules so the loop converges, ends, or escalates to a human instead of burning tokens on two agents disagreeing. Does not write

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution.

Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.

Guide for adding new AI provider packages to the AI SDK. Use when creating a new @ai-sdk/<provider> package to integrate an AI service into the SDK.

建立與更新符合專案最新規範（如 description 換行與觸發詞語法、title 欄位使用時機、標籤正規化及 references 目錄規劃）的高品質技能之微調版指南。 當使用者提及以下關鍵字或情境時觸發： - "建立新技能" 或 "建立技能" - "微調技能" 或 "微調技能指南" - "新增技能 / create skill" - "技能 Frontmatter 規範 / skill frontmatter" - "技能觸發詞 / skill triggers"

The core OpenAI-compatible inference endpoints: chat completions, embeddings, images, audio (TTS/STT), moderations, rerank, and the Responses API. The primary integration surface for AI agents.

Convert various file formats (PDF, Office documents, images, audio, web content, structured data) to Markdown optimized for LLM processing. Use when converting documents to markdown, extracting text from PDFs/Office files, transcribing audio, performing OCR on images, extracting YouTube transcripts, or processing batches of files. Supports 20+ formats including DOCX, XLSX, PPTX, PDF, HTML, EPUB, CSV, JSON, images with OCR, and audio with transcription.

說明如何在 OpenCode 平台中取得工具定義（Tool Definitions）。AI 可透過會話初始化時自動注入的 System Prompt 取得工具定義的完整內容，包括工具名稱、參數類型與描述說明。Use when users request (1) 取得工具定義, (2) Tool definition source, (3) Tool metadata, (4) OpenCode tool definitions, (5) 工具元數據取得.

Create and debug Pi extensions with custom tools, slash commands, UI components, lifecycle event handlers, compaction hooks, and session persistence. ALWAYS use when extending Pi beyond built-ins or troubleshooting extension API behavior. Covers current @earendil-works ExtensionAPI shapes, registerCommand string arguments, registerTool execute signatures, sendMessage/sendUserMessage, ctx.compact(), ctx.reload(), session entries, and common migration pitfalls from older @mariozechner examples.

Builds Agent-to-Agent (A2A) servers and clients following Google's open protocol for agent interoperability. Use when the user wants to create an A2A-compliant agent, build an Agent Card, implement task management, connect agents across frameworks, set up agent discovery, handle streaming responses, implement push notifications, or orchestrate multi-agent workflows. Trigger words: a2a, agent to agent, agent2agent, a2a protocol, a2a server, a2a client, agent card, agent interoperability, agent co

This skill should be used when the harness, scaffold, workflow, or optimizer itself is the optimization target: recursive self-improvement (RSI) loops, meta-harnesses, self-improving harnesses that mine their own failures and propose bounded edits, evolutionary or population-based search over agent scaffolds, acceptance gates for self-modifying systems, and agentic context evolution where the mechanism that produces context is versioned and evolved. Route governance of a single autonomous loop (

Design optimal agent team compositions with sizing heuristics, preset configurations, and agent type selection. Use this skill when deciding how many agents to spawn for a task, when choosing between a review team versus a feature team versus a debug team, when selecting the correct subagent_type for each role to ensure agents have the tools they need, when configuring display modes (tmux, iTerm2, in-process) for a CI or local environment, or when building a custom team composition for a non-sta

Acquire a training data source with license validation and delegate ingest to the semantic memory kernel

Orchestrates the Voyager game development process.

Cohere API for enterprise NLP — embeddings, reranking, RAG, and text generation. Use when building RAG pipelines, semantic search, document reranking, or enterprise NLP applications. Command R+ excels at tool use and retrieval-augmented generation; Embed v3 and Rerank 3 are best-in-class for search quality.

Curate LLM training data: dedupe, filter, PII redaction.

Design LLM applications using LangChain 1.x and LangGraph for agents, memory, and tool integration. Use when building LangChain applications, implementing AI agents, or creating complex LLM workflows.

Call POST /embeddings on Venice. Covers request shape (input, model, encoding_format, dimensions, user), OpenAI compatibility, response compression (gzip/br), and practical usage for retrieval, clustering, and RAG.

Manage Omni Skills (list, install, test, remove), plugins (create, configure), and persistent memory (search, add, clear) from the CLI.

Research-to-implement pipeline chaining 5 MCP tools with graceful degradation

TODO(shakespii): Use when <trigger>… — third person, concrete searchable keywords; do not summarize the workflow.

Analyze prompts for clarity, structure, examples, and output reliability

Drive a real Chromium-based browser from an agent loop. Use when a task needs to navigate web pages, fill forms, scrape rendered HTML, or take screenshots.

Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.

YOLO detection, segmentation, classification, and pose estimation setup, training workflow, evaluation metrics, and XAI verification

Discover and use Venice public characters (persona-driven system prompts with a bound model). Covers GET /characters (search/filter/sort), /characters/{slug}, /characters/{slug}/reviews, the Character schema, and how to apply a character via venice_parameters.character_slug in chat completions.

Reference-grade guide to getting reliable structured output and tool/function calls out of LLMs — JSON mode vs constrained/grammar-guided decoding, response_format json_schema and tool-schema coercion, Zod/Pydantic validation at the boundary, the validate→repair→escalate→fallback loop, tool-argument validation, idempotency for safe retries, fallback chains, and evals. Use when an LLM must return parseable JSON, fill typed tool arguments, or drive deterministic downstream code.

Email infrastructure for AI agents. Create accounts, send/receive emails, manage webhooks, and check karma balance via the AgentMail API.

Build production-ready Tavily integrations with best practices baked in. Reference documentation for developers using coding assistants (Claude Code, Cursor, etc.) to implement web search, content extraction, crawling, and research in agentic workflows, RAG systems, or autonomous agents.

Guide for integrating Agentica SDK with Claude Code CLI proxy

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

Generate videos via 9Router /v1/videos/generations using xAI Grok Imagine (grok-imagine-video). Async job flow - submit, poll request_id until done, download MP4. Use when the user wants to create, generate, or render a video, text-to-video (txt2vid), or image-to-video.

You are an **Image Prompt Engineer**, an expert specialist in crafting detailed, evocative prompts for AI image generation tools. You master the art of translating visual concepts into precise, str...

WhyLabs integration skill for ML observability, profile logging, and anomaly detection.

Use when writing Java code with `dev.axllm:ax` for reward-scored generation, iterative candidate improvement, evaluator feedback, and optimizer-backed refinement patterns.

Reference for building scripted and programmatic integrations around the Antigravity CLI (`agy`) non-interactive print mode. ALWAYS use when the user wants to call `agy` from another script, server, subprocess, automation pipeline, CI job, or custom agent harness; when they mention `agy -p`, `agy --print`, `--dangerously-skip-permissions`, conversation resume, headless Antigravity usage, or Google-subscription-backed CLI automation. Covers auth reuse, subprocess patterns, workspace scoping, conv

Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid search, and distributed systems integration. Use when building distributed AI systems, multi-agent coordination, or advanced vector search applications.

Expert guidance for creating, building, and using Claude Code subagents and the Task tool. Use when working with subagents, setting up agent configurations, understanding how agents work, or using the Task tool to launch specialized agents.

Use this skill when building real-time, bidirectional streaming applications with the Gemini Live API. Covers WebSocket-based audio/video/text streaming, voice activity detection (VAD), native audio features, function calling, session management, ephemeral tokens for client-side auth,...

Discover Venice models, their capabilities, constraints, and pricing. Covers GET /models (with ?type filter), /models/traits, /models/compatibility_mapping, the ModelResponse schema (capabilities, constraints, pricing per type), and how to use this to pick the right model programmatically.

Query the ASI:One LLM — the AI model by the ASI Alliance (Fetch.ai + SingularityNET + Ocean Protocol). OpenAI-compatible API. Supports asi1 (default) and asi1-mini models. Requires ASI_ONE_API_KEY env var. Use when asked to "use ASI", "query ASI:One", or "use the Fetch.ai LLM".

Diagnoses and reduces Qdrant memory usage. Use when someone reports 'memory too high', 'RAM keeps growing', 'node crashed', 'out of memory', 'memory leak', or asks 'why is memory usage so high?', 'how to reduce RAM?'. Also use when memory doesn't match calculations, quantization didn't help, or nodes crash during recovery.

Context is the complete state available to a language model at inference time. It includes everything the model can attend to when generating responses: system instructions, tool definitions, retrieved documents, message history, and tool outputs.

Advanced swarm orchestration patterns for research, development, testing, and complex distributed workflows

Braintrust tracing for Claude Code - hook architecture, sub-agent correlation, debugging

Extract CLI command structure from commander/yargs for AI agent skill generation Use when working with agent-skills, cli, commander, documentation, skill-generation, yargs.

Jupyter notebook execution skill for running notebooks programmatically and extracting outputs.

Serverless vector database at the edge with Cloudflare Vectorize. Use when: building semantic search on Cloudflare Workers, RAG pipelines at the edge, low-latency vector similarity search, or storing and querying embeddings without managing a separate vector database.

Get your AI agent a real US/Canada phone number in one API call. Make voice calls, send and receive SMS, and hold actual conversations — all via API.

Create and train AI learning plugins with AgentDB's 9 reinforcement learning algorithms. Includes Decision Transformer, Q-Learning, SARSA, Actor-Critic, and more. Use when building self-learning agents, implementing RL, or optimizing agent behavior through experience.

Design better AI skills with proven architecture patterns. Helps you decide Workflow vs Agent, pick the right pattern (Prompt Chaining, Routing, Parallelization, Orchestrator-Workers, Evaluator-Optimizer), write clean SKILL.md files, and catch common mistakes with a 25-point quality checklist. Based on design principles from Anthropic, OpenAI, and LangChain.

Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling. Use when training massive models across multiple machines or running distributed hyperparameter sweeps.

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.

Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.

Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.

Plugin architecture specialist. Use when deciding the right component mix, structure, and metadata for a new Cursor plugin.