Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Set up the Basic Memory plugin for this project — a short guided interview that configures the project mapping, seeds note schemas, learns or suggests placement conventions, and enables capture reflexes. Use when the user runs /basic-memory:bm-setup, says "set up basic memory", or asks to configure/bootstrap the plugin.
An advanced development guide to Claude Code—a comprehensive Chinese-language tutorial covering tool usage, the REPL environment, development workflows, MCP integration, advanced patterns, and best practices. For learning Claude Code's advanced features and development techniques.
Put an idea on trial in front of a jury of 12 Claudes. A prosecutor Claude argues it fails, a defense Claude argues it works, they rebut each other, 12 juror Claudes with different backgrounds vote on their own, and a judge Claude reads the verdict and the conditions that would flip it. Use when the user wants honest pushback instead of agreement, says Claude just agrees with everything, asks "is this a good idea", wants an idea, plan, pitch or claim stress-tested, or says "court" or "put it on trial".
GitHub AI-powered security and automation features (2025-2026). Proactively activate for: (1) GitHub Copilot in PRs (Copilot code review, Copilot autofix), (2) GitHub Advanced Security (CodeQL, secret scanning, dependency review), (3) Copilot Workspace, (4) GitHub Models (gh models eval, prompt evaluations), (5) GitHub Copilot CLI 2.x, (6) AI-assisted issue triage, (7) Copilot extensions and skillsets, (8) Copilot Enterprise context (knowledge bases), (9) automated dependency updates (Dependabot AI grouping), (10) Spark for app generation. Provides a feature catalog, Copilot CLI command reference, gh models eval examples, GHAS configuration, and Copilot Workspace usage
Shape output for a reader with ADHD. Use this skill whenever responding to ANY user message including coding tasks, debugging, explanations, planning, and casual conversation. Output should lead with concrete next actions, number multi-step work, externalize state across turns, suppress tangents, give specific time estimates, and make wins visible. Trigger even on casual messages and even when the user did not explicitly ask for brevity.
Make the environmental cost of AI usage visible — floor estimates built from published per-prompt figures, rendered as lightbulb time (💡), with an optional weekly energy budget that informs but never blocks. Use when the user asks "how much energy did this session cost", "what's the environmental cost", "show the energy footer", "set an energy budget", "energy report", or enables environmental awareness in the limits-setup interview.
The AI limits interview — run by /setup-brad-frost-skills, or directly any time. A short guided interview that establishes your personal limits for AI usage — quiet hours, a sessions-per-day ceiling, how sessions end — then wires them into whatever your tools can actually enforce and tells you plainly where enforcement stops. Use when the user says "set my limits", "set up my AI limits", "limits setup", "help me stop using AI at 2am", when /setup-brad-frost-skills routes here, or when the limits skills are installed but no config exists yet at ~/.config/ai-limits/.
Bridges Claude Code and Codex sessions when native tools can't reach the target. Use for 跨产品通信 / 脚本回帖 / 送达排查: cross-product messages, hooks or scripts posting to a session, verifying delivery (Held messages included), shared-work ownership, or an inbound peer assertion. Not for same-product or parent/subagent messaging (use native tools), spawning agents, moving full history, or cross-machine pairing (deprecated).
Audit and govern Skills, plugins, and MCP servers across Codex, Claude Code, and compatible Agent hosts. Use when the user asks to inventory, deduplicate, classify, scope, archive, or reduce the context cost of Agent capabilities; asks why there are so many Skills; wants global/project/trigger decisions; or says “Skill 瘦身”, “盘点 Skill”, “整理 Skill”, “Skill 治理”, “Skill 太多了”, “哪些 Skill 在全局生效”, “分析插件和 MCP”, “上下文占用”, “生成 Skill 复审页面”, “我选好了”, or “读取我的 Skill 设置”. Default to a read-only evidence audit and a locally persisted review draft. Never install, enable, copy, archive, disable, delete, or execute a plan without separate explicit authorization for that exact phase and target set
Spawn Ori as a subprocess to run a throwaway model eval on a pinned harness and model, then relay the results. Use when the user asks which model they should use, wants to compare models, wants to measure whether their agent or prompt does the right thing, wants to catch regressions in agent behavior, or asks how good their current model is. Applies to any codebase in any language. Do not use for plain unit tests that involve no model, and do not use to re-run an eval that already exists (run `ori eval <file>` directly).
Propagate ONE validated learning about SKILL design or quality from this session into the central skill-learnings file — extract the smallest useful pattern or anti-pattern, categorize it (Structural Patterns / Content Patterns / Anti-Patterns / Skill-Specific), and add it once without touching any actual skill file. User-only maintenance workflow that mutates durable skill-authoring guidance: it grounds the rule in what THIS session revealed, never invents, checks for duplicates, and confirms category and wording before writing. Use to capture a skill-authoring rule so future skill work inherits it.
Prove a claim of done, fixed, or passing before you make it. Run the commands and read their output. Read this skill before you say that work is done, fixed, or tested. Read it also after a small change. Use it before a commit or a PR. Use it also when the user asks to write or update the verification recipe of the app.
Use when the user explicitly wants implementation planning before coding, asks to create or revise an implementation plan, supplies requirements with explicit approval evidence, a specification, acceptance criteria, or task list, or has inputs concrete enough to plan but not execute. Do not use for rough unapproved requirements drafting.
Give your AI agent eyes to see the entire internet. Install and configure upstream tools for Twitter/X, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu, Douyin, LinkedIn, Boss直聘, WeChat (微信公众号), RSS, and any web page — then call them directly. Use when: (1) setting up platform access tools for the first time, (2) checking which platforms are available, (3) user asks to configure/enable a platform channel. Triggers: "帮我配", "帮我添加", "帮我安装", "agent reach", "install channels", "configure twitter", "enable reddit"
When the user wants to deploy AI sales development reps, automate sales qualification, build signal-to-action routing, or design AI agent architecture for sales. Also use when the user mentions 'AI SDR,' 'AI sales agent,' 'automated qualification,' 'signal routing,' 'sales automation,' '11x,' 'Artisan,' 'AiSDR,' 'AI BDR,' or 'autonomous sales.' This skill covers AI SDR deployment, qualification automation, and agent architecture for sales development. Do NOT use for technical implementation, code review, or software architecture.
Deploy the OpenClaw AI agent platform on Alibaba Cloud ECS and integrate it with a DingTalk bot. OpenClaw (formerly Clawdbot/Moltbot, Chinese name "龙虾") is an open-source AI assistant and automation platform that supports natural language-driven task automation and multi-channel chat integration. This Skill covers the full workflow from ECS instance creation, public network configuration, and base environment setup to one-click OpenClaw deployment and DingTalk bot verification. End users can chat with the AI assistant by @mentioning the bot in a DingTalk group. Triggers: "OpenClaw", "龙虾", "Clawdbot", "Moltbot", "DingTalk bot", "DingTalk AI", "deploy OpenClaw on ECS", "AI agent platform", "DingTalk integration", "openclaw dingtalk", "openclaw deploy", "DingTalk AI employee", "Alibaba Cloud OpenClaw", "Bailian + DingTalk", "DingTalk group AI", "DingTalk smart assistant", "部署龙虾", "龙虾机器人", "龙虾钉钉"
Use Bitwarden Agent Access CLI (`aac`) with agent-controlled browser login flows. Use when a user wants an AI agent to log into a website through `aac listen`/`aac run`, approve credentials in Bitwarden Agent Access, fill browser login forms without printing passwords, debug Agent Access browser login pairing, or build a credential-conscious bridge between `aac run` env injection and browser automation.
Создаёт однозначную спецификацию результата без шагов реализации. Используй, когда пользователь хочет зафиксировать задачу и контекст, целевой результат, рамки и критерии приемки, оставив технический путь исполняющему агенту. Не используй для выбора решения, составления плана реализации, непосредственной реализации или ревью изменений кода.
Design structured, engaging chatbot conversations with robust intent handling, slot filling, disambiguation, error recovery, and graceful fallback strategies. Use when the user requests chatbot conversation design or provides relevant inputs for this workflow.
When quota runs out before bed or before leaving, set a one-off local timer that automatically sends a minimal claude -p message after the Claude subscription quota resets, immediately opening a new 5-hour usage window — by the time the user wakes up, much of the new window has already elapsed, effectively gaining a free cooling period. Use when the user says "X hours until quota resets, ping Claude on a timer for me", "I'm going to sleep, ping it at the right time to open a new window", "quota is used up, ping it after the reset", or "set a timer to open a new 5-hour cooling period", or mentions a scheduled ping, pinging Claude, quota ping, opening a window, or automatically sending a message after quota reset. Limited to one-off scheduled tasks and macOS local machine (caffeinate/BSD date); for recurring tasks use /loop or cron, to check when ChatGPT/Codex resets use tibo-reset-codex, and to analyze why quota was exhausted use claude-usage-analyst.
This skill should be used for multi-session autonomous agent work requiring progress checkpointing, failure recovery, and task dependency management. Triggers on '/harness' command, or when a task involves many subtasks needing progress persistence, sleep/resume cycles across context windows, recovery from mid-task failures with partial state, or distributed work across multiple agent sessions. Synthesized from Anthropic and OpenAI engineering practices for long-running agents.
Use when the user wants to INSPECT, MONITOR, START, APPROVE, or CONTROL Archon workflow RUNS in the current project — driven through the `archon` CLI over bash. Triggers (inspect): "what's running", "list runs", "show recent runs", "run status", "did the review pass", "check run <id>", "show me run <id>", "what happened in that run". Triggers (control): "approve the plan", "approve run <id>", "reject that run", "cancel that run", "abandon run <id>", "resume run <id>", "continue that run". Triggers (start): "start <workflow> in the background", "kick off <workflow> detached". Capability: Drives `archon workflow runs/get/status/run --detach/approve/reject/cancel/abandon/resume` with machine-readable `--json` output, scoped to the current project by cwd. NOT for: Authoring workflows/commands, or Archon setup/config — use the broader `archon` skill.
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).
Production Next.js web-chat integration: tool approval, persisted conversations, tool result UI, grounded retrieval, follow-up suggestions and embedded widgets. Use when building or debugging a web chatbot that needs these features. Use ai-app for scaffolding and ai-sdk for general SDK APIs; multi-platform bots use Chat SDK
(NS) LangGraph.js agent-api — bootstrap-agent-runtime, StateGraph, MCP, skill bind/inject, checkpointers, user memory, HITL/SSE. Use for new agent-api, LangGraph graphs, MCP wiring, orphan layout, bind parity, graph-spec sync, or “fix my LangGraph agent” / “wire MCP tools” / “translations in the graph.” Do NOT use for CrewAI, generic web apps, or SDD-only with no agent-api.
Multi-channel marketing orchestrator for campaign planning, specialist routing, and marketing coordination. Use when the user needs a marketing plan, brand strategy, campaign coordination, or help deciding which marketing specialist to use. Triggers: 'marketing plan', 'brand strategy', 'campaign management', 'plan a campaign', 'coordinate marketing', 'which marketing agent should I use'.
Find, connect, and use MCP tools and skills via the Smithery CLI. Use when the user searches for new tools or skills, wants to discover integrations, connect to an MCP, install a skill, or wants to interact with an external service (email, Slack, Discord, GitHub, Jira, Notion, databases, cloud APIs, monitoring, etc.).
When the user is a solo founder building their GTM motion, wants to scale without hiring, or needs to design an AI agent team for go-to-market. Also use when the user mentions 'solo founder,' 'one-person startup,' 'solopreneur,' 'bootstrapped,' 'no team,' 'AI agents as team,' 'scaling without hiring,' 'founder-led sales,' 'lean GTM,' 'one-person company,' or 'no employees.' This skill covers the complete solo founder GTM playbook from stack selection through agent team design, revenue-stage transitions, time allocation, and when to finally hire. Do NOT use for technical implementation, code review, or software architecture.
Generate tailored AI agent instruction files using the AgentRC instructions command. Produces .github/copilot-instructions.md by default (recommended for Copilot in VS Code), plus optional per-area .instructions.md files with applyTo globs for monorepos. Use after running /acreadiness-assess to close gaps in the AI Tooling pillar.
Salesforce Agentforce AI agents and autonomous automation (2025-2026). PROACTIVELY activate for: (1) building Agentforce agents and topics, (2) Agent Builder configuration, (3) Atlas Reasoning Engine usage, (4) Agentforce action setup (Apex, Flow, Prompt Templates), (5) topic-based routing and instructions, (6) Agentforce Service Agent vs Sales Agent vs Custom, (7) Einstein Trust Layer (data masking, toxicity detection), (8) prompt engineering for grounded responses, (9) Agentforce testing and Test Center, (10) deploying agents across orgs. Provides: agent design patterns, topic and action templates, prompt engineering recipes, Trust Layer configuration, and deployment workflow
When the user wants to deploy AI sales development reps, automate sales qualification, build signal-to-action routing, or design AI agent architecture for sales. Also use when the user mentions “AI SDR,” “AI sales agent,” “automated qualification,” “signal routing,” “sales automation,” “11x,” “Artisan,” “AiSDR,” “AI BDR,” or “autonomous sales.” This skill covers AI SDR deployment, qualification automation, and agent architecture for sales development.
Generate, refine, and validate Character Card V3 cards and lorebooks. Use when creating new characters from a concept, converting or upgrading existing cards to CCv3, improving weak character behavior consistency, or building lorebook entries that steer scenario outcomes while staying token-efficient.
Generate and optimize CLAUDE.md files for Claude Code projects. Capabilities: initialize project instructions, analyze codebase context, optimize existing CLAUDE.md, apply Anthropic best practices, reduce token usage, and improve effectiveness. Actions: initialize, create, optimize, and enhance CLAUDE.md files. Keywords: CLAUDE.md, project instructions, Claude Code setup, project context, codebase analysis, Anthropic best practices, token optimization, project configuration, AI instructions, coding guidelines, project rules, workspace setup. Use when initializing CLAUDE.md for new projects, optimizing existing project instructions, setting up Claude Code for a codebase, or improving AI coding guidelines.
Use when about to consume content from any external/untrusted source — files from externally-cloned repos, WebSearch results, WebFetch responses, or fetched files. Identifies and neutralizes prompt-injection attempts via hybrid regex + LLM detection. Returns sanitized content plus a structured report. Maintains a persistent flagged-sources document at docs/security/flagged-sources.md that turns repeat-offender sources into automatic caution-bumps. Severity-keyed action: low and medium are removed (replaced with [REMOVED: marker]), high aborts the whole sanitization. Flagged content is never echoed back to the caller. One layer of defense in depth — combine with synthesis guard rails and user review. Hand-authored, security-critical.
Execute any of 240+ specialized prompt patterns natively across Extraction, Summarization, Analysis, Creation, Improvement, Security, Rating. Common: extract_wisdom, create_threat_model, analyze_claims, improve_writing, review_code, mermaid, youtube_summary. CLI used only for YouTube transcript (-y) and URL fallback (-u). Two workflows: ExecutePattern, UpdatePatterns. USE WHEN fabric, fabric pattern, run fabric, update patterns, threat model, analyze claims, improve writing, review code, mermaid, STRIDE, sigma rules. NOT FOR multi-agent investigation (Research) or content-adaptive extraction (ExtractWisdom).
Use when designing hosted or background agent infrastructure: sandboxed execution, remote coding environments, warm pools, session persistence, multiplayer collaboration, self-spawning agents, or Modal-style sandboxes.
Honor the quiet hours the user set for themselves — the hours where AI tools are off-limits so "one more prompt" stops costing sleep. Enforced as refused tool calls where hooks exist, advisory everywhere else, with a deliberate logged override path. Use when the user asks about their quiet hours, asks why tool calls are paused, wants to override quiet hours, or asks "am I inside my quiet hours".
Enforce the sessions-per-day ceiling the user set for themselves — count sessions honestly, refuse new ones past the limit with the same terse message every time, and make raising the limit a deliberate, logged act. Use when the user asks "how many sessions have I used", "am I at my limit", "raise my session limit", "limits report", or asks why a session was refused.
The always-on protocol for the loam skill namespace. Use at session start and whenever a loam task appears. Routes goals and other loam work, explains the memory model (memory = umbrella; wiki, guidance, and checkpoints are substrates), and lists cross-cutting rules. This is a routing/meta skill — delegate to a specific loam skill rather than performing work itself.
Periodic pass that keeps a project's verification skill and feature map honest: parallel source readers per feature, one live session driving every feature, at most one PR of proven corrections. Use for /maintain-verification-skill or "audit the verify skill".
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.