Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Drive Roblox Studio from an AI agent via the built-in Studio MCP server - connecting Claude Code, Codex, Cursor, OpenCode, Gemini CLI; tools to explore the DataModel, read/edit scripts, run Luau in edit/server/client, playtest, read output, capture screenshots, simulate input, insert assets. Use when an agent should inspect, modify, or playtest an open Studio place, or to set up the connection.
EdgeOne Makers full-stack development skill pack — covers AI agent development (DeepAgents、LangGraph、 Claude SDK、OpenAI Agents、CrewAI), cloud functions (Node.js/Go/Python), edge functions, KV storage, middleware, and rapid deployment, helping AI coding assistants build and release applications on the EdgeOne platform accurately and efficiently.
Mem0 provider for Vercel AI SDK (@mem0/vercel-ai-provider). TRIGGER when: user mentions "vercel ai sdk", "@mem0/vercel-ai-provider", "createMem0", "retrieveMemories", "addMemories", "getMemories", "searchMemories", "mem0 vercel", "AI SDK provider", "AI SDK memory", or is using generateText/streamText with mem0. Also triggers for Next.js apps needing memory-augmented AI. DO NOT TRIGGER when: user asks about direct Python/TS SDK calls without Vercel (use mem0 skill), or CLI terminal commands (use mem0-cli skill).
Use when a person in their own Claude Code or Codex session mentions OpenRig, rigs or the rig command, or wants a team of coding agents (Claude Code, Codex) on their project. That includes installing, setting up, starting or updating OpenRig, seeing, getting back to or re-attaching to their OpenRig agents (including bringing a stopped team back), TUI or operator, starting or sharing a team, and joining this session to one. Operating a running rig from inside it (fleet recovery, seat handover, watchdog wakes, in-rig upgrades) is the openrig-skills skill.
Applies this repository's skill-authoring standard as a procedure. Use for any change to, or judgment about, a file under skills/** — a SKILL.md, a reference, a persona prompt, a bundled script's instructions: creating a skill, editing one, reviewing a skill change, or acting on review feedback (human or bot) about one. Not for src/, tests/, or scripts/ code.
Meta-skill for making the agent self-improving. Covers updating AGENTS.md, creating new skills from repeated workflows, and deciding what to systematize. Invoke after completing tasks, when noticing repeated friction, or at session end.
Oracle Cloud Infrastructure guidance for designing, operating, and troubleshooting OCI services, including OCI Kubernetes Engine (OKE), OCI Internet of Things Platform, OCI Functions deployment and troubleshooting, and Enterprise AI workflows for OCI Generative AI models, Responses API agents, RAG, cost estimation, governance, private endpoints, hosted agentic applications, and Oracle platform integrations. Use when the user asks about OKE cluster design, Terraform or Resource Manager planning, OKE incident troubleshooting, Generic VNIC Attachment, Multus, pod networking, node pools, add-ons, ingress, load balancers, OCIR image pulls, Workload Identity, Kubernetes workloads on OCI, OCI IoT domains or digital twins, device publish flows, OCI Functions setup, deployment, invocation, or troubleshooting, OCI Generative AI, Enterprise AI Models, Enterprise AI Agents, governed GenAI applications, agentic workflows, RAG on Oracle Cloud, or OCI Generative AI pricing.
Recall and persist long-term memory through the OpenViking MCP tools. Use at the start of any substantive task (coding, configuration, debugging, multi-step or tool-based work) to retrieve relevant prior knowledge with find/search/read, and during or after work to persist durable facts, preferences, decisions, and lessons with remember. Do not use for casual chat or simple factual questions the model can answer directly.
Integrate Shengwang products: ConvoAI voice agents, RTC audio/video, RTM messaging, Cloud Recording, and token generation. Use when the user mentions Shengwang, 声网, ConvoAI, RTC, RTM, voice agent, AI agent, video call, live streaming, recording, token, or any Shengwang product task.
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or the loop cap is reached. Use when: implementing features from a spec or plan, hardening existing code, fixing a batch of issues, or any multi-file task that benefits from a build-verify-fix cycle.
Caveman mode in Spanish. Cuts ~75% of tokens by speaking like a technical caveman. Same technical precision, less wordiness. Levels: lite, full (default), ultra. Use when the user says "caveman mode", "talk like a caveman", "less tokens", "be brief", or invokes /caveman-es.
Claude Code Command Selection Guide - Automatically recommend and select the right commands, agents, and skills in Claude Code. Use when: (1) user is unsure which command or tool to use, (2) needs to decide which agent/skill best fits the current task, (3) querying usage scenarios for /plan, /tdd, /compact, /loop and other commands, (4) understanding when to invoke planner, code-reviewer, build-error-resolver and other agents, (5) needs command cheat sheet or decision flowchart. Triggers: "which command to use", "which agent", "command selection", "how to use /plan", "when to use /compact", "agent selection guide", "command cheat sheet", "skill recommendation".
Use this skill when writing commands, hooks, skills for Agent, or prompts for subagents or any other LLM interaction, including optimizing prompts, improving LLM outputs, or designing production prompt templates.
A skill for improving prompts by applying general LLM/agent best practices. When the user provides a prompt, this skill outputs an improved version, identifies missing information, and provides specific improvement points. Use when the user asks to "improve this prompt", "review this prompt", or "make this prompt better".
Orchestrates implementation of a plan file by delegating work to subagents in parallel. Verifies git branch state, tracks progress, and ensures high-quality implementation. Invoke with a plan file path and optional model override: /implement plans/my-plan.md [--model sonnet]
Use this skill when the user asks to "create a hedera policy," "block a tool," "limit recipients," "extend AbstractPolicy," "reject a tool," "enforce a threshold," "rate limit a tool," "block transfers over X," "allowlist tools," or needs guidance on writing blocking validation rules for Hedera Agent Kit tools. Policies block tool execution by throwing — for non-blocking observation use the hook skill instead.
Provides image recognition for non-multimodal models (purely text models such as deepseek-v4-pro, GLM-5.1, mimo-v2.5-pro). Triggered automatically when the main model cannot recognize images, when the user sends a screenshot/design draft/UI screenshot that needs analysis, or when the user says "look at this image", "analyze this screenshot", "what's wrong with this image". Also applies to any case where the user pastes an image but the current model does not support image input. Supports recognizing multiple images at once, with primary/backup fallback via multiple configured vision models. Can also be triggered manually with the commands /skill:vision-support or /vision. Hard rule: the models configured for this skill are used only for image content recognition and never take part in the main logical reasoning. Note: if the current model is itself multimodal (such as Claude Sonnet 4, GPT-4o, Gemini, which can recognize images directly), do not use this skill — just let the main model do the recognition.
Set a tracking document as the source of truth for the current feature or task. Use when starting work on a feature, bug fix, or multi-step task that benefits from a persistent record of decisions, discoveries, and progress. Keeps the document updated as work proceeds.
This skill should be used when the user asks to “migrate to hedera-agent-kit v4”, “upgrade hak v3 to v4”, “convert plugin to BaseTool”, “migrate Tool to BaseTool”, “rename hedera-agent-kit imports”, “fix v3 deprecation warnings”, or has a project still on `[email protected]` / `@hashgraph/sdk` that needs to move to `@hashgraph/[email protected]` / `@hiero-ledger/sdk`. Covers both custom-plugin migration (Tool → BaseTool) and toolkit-consumer migration (LangChain/AI SDK/ElizaOS/MCP).
This skill should be used when the user asks to “create a hedera hook”, “write an agent kit hook”, “extend AbstractHook”, “add a tool lifecycle hook”, “log tool executions”, “audit trail hook”, “track tool calls”, or needs guidance on building observation/logging/metrics extensions for Hedera Agent Kit tools. Hooks are non-blocking — for blocking validation rules use the policy skill instead.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.