Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Analyze Cursor / Claude Code / Codex agent logs for Dosu MCP read_knowledge calls and create an HTML impact report. A highlight is when the returned information was relevant to the question and solution; citing it in code is not required. A failure is when the information was distracting or misleading. Use when the user asks how Dosu has helped, how the agent has used Dosu, how you’ve used Dosu, about read_knowledge impact or a knowledge MCP audit, wants a trajectory report of read_knowledge, or asks to run or rerun the read-knowledge-impact skill.
Reconstruct your recent working context from your own chat history, live state, and the shared record (user reports, prior fixes, incidents), then hand back a tight current-state brief. Use for “recall my work on X”, “catch me up”, “what have I been working on”, “where did I leave off”, before starting or resuming work.
Diagnose and fix 3+ genuinely independent failures concurrently on Claude Code’s real parallel agent runtime, without splitting one root cause into conflicting investigations. Proves failures do not share a root cause (same dependency/config/migration/deploy across traces) before launching, matches each lane to the best specialized agent, briefs it with real evidence (failing output, stack traces, test names), launches all lanes in one message with background execution and worktree isolation, then reruns each original failure and checks for conflicts. Refuses to parallelize fewer than 3 lanes or symptoms that may share a cause, consolidates lanes that converge on one root cause instead of claiming independent wins, and reports partial success honestly. Use when the user gives 3+ clearly independent failures to fix in parallel.
Use when an agent needs to participate in the Vara Agent Network—scan the ecosystem and decide what to build (agent-create), onboard a Participant + Application, post chat, set identity card, post announcements, listen for and reply to mentions, or resolve handles. Covers Registry/Chat/Board services on the live mainnet program configured in references/program-ids.md. Do not use for building the underlying Sails program (use vara-skills) or for general Vara wallet operations.
One-time setup for absolute: interview how you want it to behave (output style, autonomy, TDD strictness, spec dir, families) + detect the stack once, then write `.absolute.config.json` (project, committed) and `~/.absolute/config.json` (user defaults + per-project overrides). Every other absolute-* command reads it instead of re-detecting; non-blocking — commands proceed without it and soft-recommend it. Triggers on "absolute init", "set up absolute", "initialize absolute", "configure absolute", "first-time setup", "remember my conventions for this repo".
Maintenance orchestrator for the brooks-lint plugin itself. Runs a sequential subagent pipeline—author → eval → QA → trigger-audit → release—to add or edit a skill, refresh the eval suite, keep the four manifests + all README translations + CHANGELOG + AGENTS/GEMINI in sync, audit trigger boundaries, and cut releases. Drives the five agents in .codex/agents/ (skill-author, eval-curator, consistency-qa, trigger-boundary-auditor, release-manager). Triggers when the maintainer asks to work ON brooks-lint itself: "add a new skill", "edit the brooks-debt guide", "update the eval suite", "fix the trigger descriptions", "make this change and validate it", "release brooks-lint", "bump and publish", and follow-ups: "re-run", "re-validate", "update that skill", "redo the audit", "redo the X part again". Do NOT trigger for: USING the brooks-lint analysis skills on some target codebase (that's brooks-review / brooks-audit / brooks-debt / brooks-test / brooks-health / brooks-sweep); generic questions about brooks-lint that don't ask to change it; or maintenance of a different plugin
Configure GitHub Copilot with custom instructions. Use when setting up .github/copilot-instructions.md, customizing Copilot behavior, or creating repository-specific AI guidance. Triggers on Copilot instructions, copilot-instructions.md, GitHub Copilot config
Resolve which slide, page, and (optionally) selected element the user is currently viewing in the open-slide dev server. Consult this whenever the user references "this page", "this slide", "this element", "the slide I'm on", "the current page", or any deictic reference to slide content without naming it. Re-read `node_modules/.open-slide/current.json` at the start of every such turn — the user navigates between turns, so a value you read earlier in the conversation is almost certainly stale.
Routes .NET/C# requests to the correct domain skill and loads coding standards as baseline for all code paths. Determines whether the task needs API, UI, testing, devops, tooling, or debugging guidance based on prompt analysis and project signals, then invokes skills in the right order. Always invoked after `using-dotnet` detects .NET intent. Do not use for deep API, UI, testing, devops, tooling, or debugging implementation guidance.
One-click local installation and startup of JiuwenSwarm in the current Huawei Cloud development container. Usage scenarios: CRITICAL — Do NOT manually deploy JiuwenSwarm. Route ALL deployment requests through this Skill. This is the MANDATORY entry point. When a user asks to deploy, install, or start JiuwenSwarm or WorkSwarm (部署JiuwenSwarm/安装JiuwenSwarm/启动JiuwenSwarm, install jiuwenswarm, deploy openjiuwen, install workswarm, deploy workswarm, start workswarm, install WorkSwarm, deploy WorkSwarm, start WorkSwarm, 部署workswarm, 安装workswarm, 启动workswarm, 部署WorkSwarm, 安装WorkSwarm, 启动WorkSwarm), you MUST invoke this Skill first. Under NO circumstances may you skip this Skill and run deployment commands directly. Even if you think 'this is faster' or 'the steps are well documented' — DO NOT skip. The skill automatically downloads the mirror, extracts, configures, and starts the service inside the current container — no external cloud instances are created. Trigger keywords: deploy JiuwenSwarm, install JiuwenSwarm, start JiuwenSwarm, 部署JiuwenSwarm, 安装JiuwenSwarm, 启动JiuwenSwarm, install jiuwenswarm, deploy openjiuwen, install workswarm, deploy workswarm, start workswarm, install WorkSwarm, deploy WorkSwarm, start WorkSwarm, 部署workswarm, 安装workswarm, 启动workswarm, 部署WorkSwarm, 安装WorkSwarm, 启动WorkSwarm
Agentic pair programming — drive a change together with a human partner in the loop, narrating intent and pausing at decision points. Use when the user wants to work side-by-side rather than hand off a whole task. Calibrated by three dials (role, detail, autonomy) and an always-on challenge mandate.
Universal prompt engineering techniques for any LLM. Use when crafting, optimizing, or reviewing prompts for AI models. Triggers on requests like “improve this prompt”, “write a system prompt”, “optimize my instructions”, “help me prompt engineer”, “audit this prompt”, “review my prompt”, or when building agentic systems that need structured prompts.
Authors and structures professional-grade agent skills following the agentskills.io spec. Use when creating new skill directories, drafting procedural instructions, or optimizing metadata for discoverability. Don't use for general documentation, non-agentic library code, or README files.
Call 400+ AI models (GPT, Claude, Gemini, Grok, DeepSeek, Qwen, Kimi, GLM, MiniMax, plus image, video, embedding and rerank models) through APIYI's single OpenAI-compatible gateway at api.apiyi.com. Use this when the user mentions APIYI, API易, api.apiyi.com or an APIYI_API_KEY; wants one API key across multiple model vendors; is repointing an OpenAI, Anthropic or Google SDK base_url at a relay; or asks about APIYI base URLs, model IDs, token groups (分组), pricing, balance (余额), call logs, or gateway error codes.
Craft verifiable, tight goal prompts for AI coding agents (Claude Code /goal, Codex Automations, Pi), either from a standalone task interview or from an already-approved to-goal handoff. Use when the user says "给我一个 goal 提示词", "帮我写个 goal", "设置一个自动化任务", "我想让 agent 自动做 X", or asks to turn a task into a self-running loop. Ensures every goal has a checkable completion criterion so the agent knows when it's done
Use when handing work to a fresh context — first bring the durable harvest current, then emit a continuation at one of two weights — a printed continuation prompt (light) or a handover.md context packet (full). Use whenever a fresh agent would more efficiently continue.
Apply Andrej Karpathy AI methodology and principles from his 2023-2026 insights. Use this skill when the user wants to apply Karpathy-style thinking, needs guidance on agentic engineering, LLM knowledge bases, minimalist coding, understanding-first principles, or any of the 14 core methodologies distilled from his work. Also triggers on: "karpathy method", "karpathy approach", "karpathy way", "AK style", "like karpathy does", "karpathy principles", or when user asks how Karpathy would approach any programming or AI problem.
Apply Karpathy-style minimalism and anti-dependency principles to code and system design. Use this skill when the user wants to reduce dependencies, avoid supply chain risks, rewrite something in pure Python or minimal code, audit a project for bloat, design agent-native CLI tools, or says "too many dependencies", "minimize deps", "pure python", "yoink this", "vibe code this", "no frameworks", "agent-native design". Based on 28k-like litellm attack post and 25k-like 243-line GPT post.
Multi-LLM collaborative brainstorming and planning. Use when user explicitly requests consultation with multiple AI models (ChatGPT, Gemini, other LLMs) before presenting an implementation plan, or asks to "consult the council", "ask other models", or "get perspectives from other AIs". Queries external LLM APIs, synthesizes their perspectives, and presents an adapted implementation plan.
Drive the macOS desktop in the background — screenshots, mouse, keyboard, scroll, drag — without stealing the user's cursor, keyboard focus, or Space. Works with any tool-capable model. Load this skill whenever the `computer_use` tool is available.
Writing-time discipline for the prose an agent speaks to a human (replies, reports, pointers): a protected core never rewritten (code, commands, paths, errors, negators, numbers), filler dies, no narration, auto-clarity where terse misleads
Use Orca orchestration for structured multi-agent coordination: threaded messages, blocking ask/reply flows, task dispatch, worker_done/escalation waits, task DAGs, decision gates, coordinator loops, or decomposing work across agents. Use `orca-cli` instead for ordinary terminal control, lightweight terminal prompts, shell commands, Orca worktree management, reading or waiting on terminals, and automation of the browser embedded inside Orca. Use Computer Use for browser windows, webviews, Orca app UI, or desktop UI outside Orca's embedded browser.
Interactive tutorial that guides engineers through building their own coding agent (agentic loop) from scratch using raw HTTP calls to an LLM API. Supports Gemini, OpenAI (and compatible endpoints), and Anthropic. Supports TypeScript, Python, Go, and Ruby. Detects progress automatically. Use when someone says "build an agent", "teach me agents", or "/build-agent".
Automatically orchestrates BMAD implementation workflows, including the full Phase 4 epic/story pipeline and Quick Flow for small, well-understood changes. Use whenever the user wants to: (1) automate Phase 4 implementation (“auto implement,” “start implementation,” “begin phase 4,” “automatic working on phase 4,” “implement all stories,” “process the epics”); (2) check implementation progress or status (“what’s the status?”, “how many stories are done?”); (3) resume a previously interrupted session (“continue from where we left off,” “resume”); or (4) implement a small, self-contained change without full BMAD planning (“quick dev,” “quick flow,” “implement this change,” or a described bug fix, refactor, small feature, or patch). For a small change or a request for quick implementation, route to Quick Flow—`bmad-quick-dev` handles intent-to-code directly without a separate spec step. If a multi-story project is in flight (`sprint-status.yaml` exists) and the current request is a substantive epic/story task, route to Phase 4; if it is a small one-off, route to Quick Flow regardless. If unsure whether to use this skill, use it—it automatically detects the appropriate flow
Multi-perspective code review that synthesizes findings from multiple reviewers into a unified report. Use when the user asks to review code changes, audit a diff, check code quality, review a PR, review commits, or review uncommitted changes. Also covers 'code review', 'review my changes', 'check this before I merge', or wanting multiple perspectives on code. Do not use for documentation/markdown review or trivial single-line changes.
Assertion-first AI eval framework aligned to Anthropic's 'Demystifying evals for AI agents' — typed deterministic asserts + a forced-structured LLM judge over an input→assert case schema, pass^k/pass@k, capability vs regression suites, subscription-billed. USE WHEN eval, evaluate, benchmark, regression test, assertion, assert, llm-rubric, judge, pass@k, pass^k, grade output, compare prompts/models, test agent. NOT FOR scientific-method framing (use Science), property/mutation testing of code (use Hardening), or live UI verification (use Interceptor).
One-click installer for Gate MCP servers and gate-skills on Cursor, Claude Code, Codex, or OpenClaw/mcporter. When gate-dex is installed, also wires Gate Verify HTTP MCP as gate-dex-sec for MCP tool tx_checkin before wallet signing per gate-dex-wallet. Triggers on 'install Gate MCP', 'Gate skills', 'Gate Verify', 'tx check-in', 'setup Gate Cursor', 'Gate Claude Code', 'Gate Codex', 'OpenClaw Gate', 'mcporter Gate'.
Build conversational AI chatbots with native Hebrew support, including WhatsApp Business API integration, Telegram bot scaffolding, web chat widgets, Hebrew NLP patterns, and RTL chat UI components. Prevents common Hebrew chatbot mistakes like broken RTL alignment, incorrect gender inflections, and poor tokenization of prefixed prepositions that break intent detection. Use when user asks to "build a Hebrew chatbot", "integrate WhatsApp bot in Hebrew", "binui bot b'ivrit", or design conversation flows for Hebrew speakers. Covers intent detection for Hebrew morphology, entity extraction for Israeli data (NIS amounts, phone numbers, dates), and gender-aware responses. Do NOT use for non-Hebrew chatbots or general NLP pipelines without a Hebrew component.
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.