Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Use when writing a new Agent Skill, editing an existing SKILL.md, or getting a skill change actually deployed — where the source lives versus the installed copy, why editing the installed copy silently does nothing, the edit → gate → commit → push → re-install → verify sequence, how to publish a skill repo so `skills add` finds it, how to word a `description` so it triggers on real requests without stealing another skill's, and when something should be an instructions-file rule instead of a skill at all.
Packages the user's agent skills into individual zips for easy upload to platforms. The default source is the current user's ~/.agents/skills, one <name>.zip per skill (keeping the top-level <name>/ directory inside the zip), with a MANIFEST.txt attached. User-invoked; the model does not trigger it automatically.
Create new skills, improve existing skills, and understand skill best practices for Builder.io Fusion. Use when the user wants to create a skill, build a slash command, automate a workflow, add custom AI instructions, create reusable AI behaviors, or asks how to extend Fusion's capabilities in their project. Also use when the user wants to improve, audit, or troubleshoot an existing skill's triggering or behavior.
Premium Skill Creator by Kevin Jeppesen (The Operator Vault). Create better OpenClaw skills with a premium first-use setup wizard pattern, minimal context bloat, and reusable scaffolding. Links: YouTube https://www.youtube.com/@kevin-jeppesen | Skool https://skool.com/operator-vault | Site https://theoperatorvault.io | X https://x.com/seo_ecom | LinkedIn https://www.linkedin.com/in/kevin-jeppesen/ | Facebook https://www.facebook.com/kevinjeppesen/
Use when the user wants to measure or set up evals/checks for one of their skills — how fast it is, whether its output is valid, whether it fires when expected, or whether its opening classification/routing gate labels inputs correctly.
Convert Claude Code skills to work on OpenAI Codex, Google Gemini CLI, Google Antigravity, and Cursor. Analyzes platform-specific features, generates target files (openai.yaml, AGENTS.md, GEMINI.md, .mdc rules), adapts frontmatter, converts MCP config, and produces compatibility reports. Use when user says "convert skill", "port skill", "multi-platform", "skill for codex", "skill for gemini", "skill for antigravity", "skill for cursor", "cross-platform skill", "convert to codex", "convert to gemini", "convert to antigravity", or "convert to cursor".
Comprehensive guide for discovering, understanding, and installing Claude Code skills from skills.sh marketplace. Use this skill when users want to (1) browse or search skills.sh for recommendations, (2) get skill recommendations by role (developer, marketer, designer, product manager), (3) understand how a specific installed skill works by reading and explaining its SKILL.md, (4) learn the anatomy and structure of skills, or (5) find and install skills that match their specific workflow needs. This is NOT for creating new skills (use skill-creator instead).
Create custom slash commands for Claude Code including syntax, arguments, bash execution, file references, and frontmatter configuration. Use when creating slash commands, custom commands, .md command files, or when asked about command creation, /command syntax, or command best practices.
Orchestrates a multi-module design session — prereq self-check, frontend routing stamp, issue clustering, module/seam design via codebase-design, TDD-ready issue rewrite, and batch tracker mutation — for a PRD + published backlog that spans two or more modules. Use when module and seam choices are still implicit after /to-issues has run.
Run the AI team pipeline for squad tasks — orchestration loop with 2 agents (Worker, Reviewer), single-step execution, and code review. Use /squad-run to execute tasks through the 7-column pipeline. AUTO-TRIGGER when: user says "implement task NNN" or any task ID + implement/build/do combination; or user confirms with "yes/ok/go/do it" after Claude proposes implementing a specific squad task.
Iteratively build multi-page websites using Stitch. Reads next-prompt.md (the baton), generates the next page with Stitch MCP, integrates it into the site, then updates next-prompt.md to continue the loop. Works with stitch-design-md and stitch-ui-prompt-architect for consistent multi-page output.
Flip Claude's default from validation to adversarial critique. Use when you are about to make a high-stakes decision, commit to a plan, or pitch something you have not stress-tested. Produces structured challenges, steelmanned counter-arguments, and the strongest case against your position — a genuine thinking partner, not a mirror.
TeamLead orchestration — use when a Human wants coordinated implementation/delegation for product code, tests, config, infra, features, bug fixes, refactoring, or behavior/UI/API changes. Do not trigger for read-only inspection/status/explanation/lightweight summary/no-edit review/research unless the Human explicitly requests TeamLead orchestration or delegation would clearly help. Work that creates or modifies behavior must pass the Research Gate (spawn research agents immediately after classification) and always invoke `superpowers:brainstorming` first, then the Lead performs the Spec/Plan gates before delegating; the Lead must not write product code themselves. Every spawn must pick a spawn mode (one-shot Agent vs teammate) and specify the model per the Model Policy (Fable only for thinking/planning work, Opus for Dev/QA/research). Execute in batches per the Commit Plan, run tests per the fast/heavy policy (full suite + E2E only once at the end), Human checkpoints only at 3 points, and every prompt includes the repo's Convention Anchor.
Use when creating or editing skills, before deployment, to verify they work under pressure and resist rationalization. Applies the RED-GREEN-REFACTOR cycle to process documentation by running a baseline without the skill, writing to address failures, and iterating to close loopholes.
The single front door to the SDD pipeline (docs/plans/): research → plan → workshop → ADR → phase tasks → implement → progress → review → ship. Use when the user wants to plan, research, explore, specify or architect (now one atomic 'plan' step → one document: business spec + implementation plan), clarify, workshop, write an ADR, break a phase into tasks, implement or build a phase (optionally with a live review companion), update progress, code-review, or ship a plan (push + PR + watch checks) — or types /the-flow (or a legacy /plan-N command), or asks to start, resume, or adopt a plan flow. Guided mode (no args) coaches from durable on-disk state; direct jump runs one stage: /the-flow <id> <verb> [flags] — 1a explore, 1b plan (spec + plan in one doc; incl. clarify), 2c workshop, 3a adr, 5 tasks, 6 implement (--companion for live review), 6a progress, 7 review, 8 ship, 8c reconcile. Ids and verbs resolve alone; printed commands carry both.
Verify whether an AI agent's decision is well-reasoned before executing. Adversarial multi-model critique (Claude, Grok, DeepSeek) returns ALLOW, BLOCK, or UNCERTAIN with confidence score. Use before any high-value swap, trade, or autonomous action.
Trae project configuration standardization expert. Used to quickly initialize Trae project configuration files and generate project rules, user preferences and Skill templates. Use this Skill when the user needs to: (1) initialize Trae configuration for a new project, (2) generate the .trae directory structure, (3) create the USER_PREFERENCES.md user preferences file, (4) create the project_rules.md project rules file, (5) create a new Skill template.
Use when the user mentions memory search, RAG retrieval, knowledge queries, embedding rebuild, document upload, connection tokens, or invokes "/tm" / "transcendence-memory" — including indirect recall like "what did we decide last time", "上次怎么处理的", "之前的方案". Covers searching, storing, multimodal RAG queries, and troubleshooting against a self-hosted memory backend.
Use for bugs, small isolated fixes, or tiny changes that don't warrant the full architect→specify→plan→implement pipeline. Runs the direct implement→archive-ready lifecycle with implement and docs approvals, and escalates to architect if scope is larger than expected.
Plans and executes batch TYPO3 migrations and large-scale refactors across hooks, PSR-14 events, TCA, dependency injection, Fluid, namespaces, ext_localconf, Content Blocks, localization, and PHP upgrades. Use when the user asks for bulk migration, mass refactor, codemod-style TYPO3 updates, repeated fixes across an extension, or coordinated v14 modernization.
Deliver the verdict for a unit of work once the TESTER and DEV have both submitted: run the TESTER's tests against the DEV's code, arbitrate each failure against the contract (the code deviated, the test deviated, or the contract did not specify it), verify that each role touched only its own paths, and produce the traceability matrix for the PR. Use when the LEAD has received both deliveries (ENTREGA TESTER and ENTREGA DEV) for a unit of work, or when asked to verify, judge, arbitrate, or give the verdict on a team delivery
Use before claiming that work is complete, fixed, or tests have passed, and before committing or creating a PR. You must run verification commands and confirm their output before claiming success; always support claims with evidence.
Run agency-orchestrator YAML workflows directly in Claude Code / OpenClaw / Cursor — no API key required; uses the current session’s LLM as the execution engine. Trigger when the user provides a .yaml workflow file or asks for a task to be completed collaboratively by multiple roles.
Use when the user says "remind me", "in X minutes/hours", "scheduled", "daily/weekly", "set an alarm", "register project", "switch engine", "create topic". yee88 Telegram bridge tool supporting one-off reminders (+30m, +2h), scheduled tasks (cron), project management, topic management and multi-engine switching. Operated directly through yee88 CLI commands.
Bash-loop runner for unattended, fresh-context execution of ready-for-agent issues. Each iteration starts empty, reads the spec from the issue body, applies TDD, commits, advances. Use when the user wants to walk away and let a batch of ready-for-agent slices ship overnight, or asks how to install/use the AFK runner.
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.