transformer-lens-interpretability
Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-14 00:15
Agents & models
26,319 results
verl-rl-training
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms…
claude-connectors
Router skill for Claude Connectors (the user-facing MCP integration surface on claude.com), Claude Desktop extensions (MCPB packaging), MCP Apps for Claude…
agent-repo-init
One-click initialization of a multi-agent repository from the Antigravity template. Use this skill when users want to scaffold a new project quickly (`quick`…
agent-design
Use when creating, editing, or reviewing a cc-suite advisor agent (a value-over-rules persona under `.cc-suite/agents/`). Covers system-prompt phrasing, model…
claude-code-conventions
Canonical reference for Claude Code plugin artifact schemas, hook events, frontmatter fields, and naming conventions. Used to inject domain knowledge into…
claude-debug
Send a bug or failing test to Claude Code for root-cause analysis and fix. Use when you've hit an error you can't trace, a test is failing for unclear reasons,…
claude-plan
Ask Claude Code to produce an implementation plan before you write code. Use when the task is complex, touches multiple files, or has unclear requirements.…
dbs-agent-migration
Agent workspace migration. Organize any project into a long-term maintainable Agent workspace that is consistent across both Claude Code and Codex: audit rule files, identify the source of truth, unify naming, and generate bridges. Trigger methods: /dbs-agent-migration, /agent migration, "migrate to Codex", "migrate to Claude Code", "unify…
init
Set up or refresh dev-team in a local Git repository, installing seven Codex agents with explicit models and drafting a reviewed repository profile.
experience-synthesis
This skill should be used when the user asks to "learn from past sessions", "extract lessons", "find what worked", "identify patterns in my workflow", "what…
writing-agents
How to write Claude Code agents that trigger reliably, use the right model, and produce consistent output. Use when creating, improving, or reviewing agent…
writing-agents
How to write Claude Code agents that trigger reliably, use the right model, and produce consistent output. Use when creating, improving, or reviewing agent…
dispatching-parallel-agents
Use when facing 2 or more tasks that can proceed independently and have no shared state or ordering dependency
executing-plans
Use when you have a written implementation plan to execute in a separate session, with review checkpoints
mcp-builder
MCP Server Construction Methodology — Systematically build production-grade MCP tools, enabling AI assistants to connect to external capabilities
workflow-runner
Run agency-orchestrator YAML workflows directly in Claude Code / OpenClaw / Cursor—no API key required, using the current session's LLM as the execution engine. Triggered when the user provides a .yaml workflow file or requests multi-role collaboration to complete a task.
coverage-gate
Enforce coverage thresholds AND test quality — coverage without behavioral assertions is meaningless.
skill-installer
Install, update, trust, or inspect DeepSeek skills from GitHub or local skill folders. Use when the user asks for available skills or wants a community skill…
openclaw-create-agent
Create and wire a new OpenClaw agent with a fixed workflow. Use when the user asks to create/add a new OpenClaw agent or says "我要创建一个新的 Agent", automate…
dataset
Search the workflow dataset for best practices from past projects, and help users contribute new run records. Each dataset entry is a real…
andrej-karpathy-perspective
Andrej Karpathy's thinking framework and expression style. Systematic distillation based on 20+ blog posts, 16 in-depth interviews, and 100+ X posts, extracting 6 core mental models, 8 decision heuristics, complete Chinese output adaptation, and a quick reference of classic phrasings. Purpose: as a thinking advisor, analyze AI technology reliability, learning methods, industry trends, and product design from Karpathy's perspective. When the user mentions "from Karpathy's perspective...
task-harness
Decompose requirements into a structured task list, and generate a task management system for long-running Agents (based on the Anthropic Effective harnesses methodology). Automatically triggered when users need to manage multi-session development tasks, track feature completion progress, or request "decompose tasks", "task management", or "project planning"
evidence-based-rag
Evidence-first retrieval-augmented reasoning skill for decision-critical scenarios. Retrieves information from a specified knowledge base, extracts verifiable…