Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Reference for the Opik SDK — tracing, span types, framework integrations, threads, and the prompt library (Python, TypeScript, REST). Use for questions like "what span types exist", "how do I flush", "track_openai", "add OpikTracer", "version a prompt". To instrument a repo end to end, use the opik-instrument skill.
Persistent file-based planning for multi-step AI-agent work. Keeps task_plan.md, findings.md, and progress.md on disk; lifecycle hooks inject selected project planning context. Automatic recovery reads project planning files only. Explicit session-catchup.py --metadata reads same-project local agent session records and emits aggregate counts only; --replay may emit bounded nonce-framed excerpts. Optional gated mode can request continuation only when the host supports it and never runs commands declared in Markdown. The skill has no network upload path. Use for research or work needing 5+ tool calls.
Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling. JSON Schema best practices, description writing that actually helps the LLM, validation, and the emerging MCP standard that's becoming the lingua franca for AI tools. Key insight: tool descriptions are more important than tool implementations.
Get a second opinion from leading AI models on code, architecture, strategy, prompting, or anything. Queries models via OpenRouter, Gemini, or OpenAI APIs. Supports single opinion, multi-model consensus, and devil's advocate patterns. Trigger with 'brains trust', 'second opinion', 'ask gemini', 'ask gpt', 'peer review', 'consult', 'challenge this', or 'devil's advocate'.
Tells your agent what Mobbin can and can't do and how the Mobbin skills work together, so it picks the right one or says when Mobbin isn't the right tool. Read it first, before any other Mobbin skill or tools, when the user asks about Mobbin or its connection, or starts a design task Mobbin could help with, including one that's too big or outside what Mobbin does
A fast Rust safety hook that blocks destructive shell, git, and filesystem commands before AI coding agents run them. 50+ security packs (databases, Kubernetes, Docker, cloud, Terraform), heredoc/inline-script scanning, CI scan mode, and explain mode; works with Claude Code, Codex, Gemini, Copilot, Cursor, and more.
Teneo CLI — 39 commands for browse & query network agents, send commands to agents, room management, deploy & manage your own agents, wallet management, daemon & updates. Query network agents, handle x402 USDC micropayments, sign on-chain transactions, auto-generate encrypted wallets, deploy your own agents with background service management (launchd/systemd). Background daemon keeps a persistent WebSocket connection. Use this skill whenever the user needs real-time Teneo data, Teneo agent payments, or Teneo agent deployment.
Configure and maintain the DeepSeek native sub-agent in the Codex desktop app, with support for model selection, routing verification, repair, deactivation, and uninstall. Use only when the user asks to configure or manage this integration; it is not triggered by ordinary API questions or by day-to-day coding tasks after configuration.
This skill helps with GitHub Copilot SDK work across Node.js/TypeScript, Python, Go, .NET, and Java. It covers setup, authentication, permissions, streaming events, custom tools, custom agents, MCP servers, hooks, skills, and session persistence.
Design and create a new hive task through guided conversation. Walks the user through problem definition, eval design, constraint specification, repo scaffolding, baseline testing with iteration, and upload. Use when user wants to create a new task, add a benchmark, or publish a challenge to the swarm.
Creative-writing domain knowledge for durable story state. Load when preserving or retrieving project memory — fact extraction, context scoping, reference writing, artifact layout, and issue tracking. If you are a knowledge agent such as kb-lead, load this for the fiction-specific categories and conventions your general methodology doesn't cover.
Objective task quality evaluation framework using quantitative KPIs. KPIs are automatically calculated by a hook when task files are modified and saved to TASK-XXX--kpi.json. Use when: reading KPI data for task evaluation, understanding quality metrics, deciding whether to iterate or approve based on data.
Drive and verify terminal applications with the termctrl CLI in a real PTY - read visible screens, run named live sessions, send typed keyboard input, wait for text, save evidence, record timelines, and export edited videos. Use when an agent must operate or test a TUI, REPL, interactive CLI, shell process, or OpenTUI application
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).
Break a one-line idea into a task brief that an AI agent can execute independently. Use it when users say "write a goal for the agent", "break down this goal in detail", "write a brief for the agent", "write a goal prompt", "let the agent run this project", or "split the work across several agents in parallel". Test in the codebase, research online when needed, ask all questions in one go (≤5), and produce a ≤4000-character brief that can be pasted into /goal and run, including measured numbers, whitelist boundaries, anti-cheat acceptance criteria, and checkpoint resumption. Execution-type and exploration-type (research/selection/solution search) tasks are routed automatically
Use when the user wants to connect an LLM or vision provider, already has an API key, asks “can I use OpenAI/Anthropic/Gemini/OpenRouter,” wants local Ollama, or needs different low-cost and powerful models. Configures provider-neutral visual understanding without tying Watch Skill to a particular agent or model vendor.
Use when the user corrects an answer about a video—“no, it actually says X,” “that's the wrong timestamp,” “you misread the error code”—or asks why an answer about a video was wrong. Records the correction as a lesson to improve future answers to similar questions and shows what the system has learned and saved.
Implements and debugs browser WebMCP integrations in JavaScript or TypeScript web apps. Use when exposing imperative tools through document.modelContext (with a navigator.modelContext fallback), annotating HTML forms for declarative tools, handling agent-invoked form flows, or validating WebMCP behavior in the current Chrome preview. Don't use for server-side MCP servers, REST tool backends, or non-browser providers.
Create detailed, execution-ready implementation plans for complex or high-risk changes without coding. Use for ExecPlan-style work, multi-hour changes, significant refactors, migrations, resumable phase checklists, and work that should be handed off to execute-plan with clear validation.
Bridge DingTalk outgoing webhook messages to Clawdbot Gateway and send replies back to DingTalk sessions. Use when setting up DingTalk as a messaging channel, troubleshooting webhook delivery, or running the local bridge service.
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.