Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Spawn a headless coding-agent CLI run for one scoped subtask. Use when the user asks to delegate work to another agent instance, run an agent CLI non-interactively (such as Antigravity's `agy -p` or Cursor's `agent -p`), automate a headless agent in a script, build, or CI step, or fan work out across isolated sessions. Do not use for interactive agent sessions or when the host's own subagent, task, or agent tool already fits.
Engage the User to resolve important open decisions in Spec Loop work, including tasks, plans, design updates, work breakdowns, implementation checkpoints, and ADR drafting. Use when the user asks to clarify, compare options, discuss criteria, or stress-test, or when different answers could materially change scope, behavior, policy, constraints, route, Design, Test specification, or ADR content. Provides mandatory decision screening for workflows that require it before dependent Spec Loop content is written or revised. Also use for general grilling when explicitly selected.
Exposes Claude's reasoning chain as an auditable, decomposable artifact. Quick mode (default) gives assumption inventory + weakest-link in 2 stages. Full mode (--full) adds decision branching, confidence decomposition, and falsification conditions. Triggers on "왜 그렇게 생각해", "reasoning", "근거", "show your work", "어떻게 그 결론이", "trace", "판단 근거", "why do you think that".
UiPath Human-in-the-Loop / HITL node authoring—building approval gates, escalations, write-back validation, and data enrichment checkpoints in Flow, Maestro, Case Management, or Coded Agents, plus a guided decision experience that captures reviewer corrections as decisions. NOT for managing, reassigning, or monitoring tasks at runtime (use uipath-tasks for that).
Transform an AI agent into a disciplined software development partner with strong judgment, transparent decisions, proportionate verification, and craftsmanship. Use for software implementation, feature work, bug fixing, refactoring, optimization, migrations, and sustained collaboration inside a codebase. Do not use merely because a task involves an AI agent; use a dedicated agent-design skill when the primary problem is agent architecture.
Use when iterating on a skill's prompt with the waxa CLI (https://github.com/mizchi/skills/tree/main/tools/waxa) — authoring scenarios, choosing graders, interpreting unclear-points, advancing a ledger, and judging convergence. Encodes the four-stage iteration pattern observed in real iter loops (structural fix → grader breadth → surface-form coverage → residual unclear) and the scenario-design pitfalls (blank-slate executor's network limit, prompt expectation explicitness, regex coverage of Japanese/English surface forms). Meta-skill: do NOT auto-invoke for routine skill edits; fires only when the user explicitly runs waxa or asks for a skill-quality eval.
After one or more discovery or analysis rounds, explicitly asks "What is still missing?" Modalities not swept, claims not verified, sources not read, edge cases not considered, downstream effects not traced, assumptions not checked. The output becomes the seed for the next round of work (more angles, more verification, more research). Use near the end of Research, Review, or Understand phases, or before declaring a task complete. Invoke via /work-gaps.
Adversarial verification ("try to refute") of a claim or a SET of claims/findings/root-causes/design-decisions using N independent sub-agent refuters per claim. Runs as a PIPELINE (each claim's refuter-group is independent — no barrier across claims) with perspective-diverse lenses as the DEFAULT for security/auth/multi-tenancy/data-integrity/correctness claims. Survival is EVIDENCE-WEIGHTED, not naive majority: only verdicts that re-read the cited source count; lazy/empty verdicts are abstentions. Strongly prefers direct deterministic oracles (commands, tests, git, type checks, exact grep) over agent voting. Use when the user says "adversarial verify", "refute this", "have verifiers attack", "perspective-diverse review", "run adversarial checks on these findings", "ultracode verify", or "kill this claim".
Parallel blind search from multiple independent angles (by structure/files, by content/semantics, by git history/time, by entity/domain, by error paths, alternative-storage-paths, config/realm dimensions, etc.), then a BARRIER at the merge step that dedups against a persistent SEEN set and lists what no angle covered. Each searcher is kept ignorant of what the others found in the same round. Use when one search strategy is known to be incomplete: "find all X", security review, "where can this break?", data flow tracing. Invoke via /work-sweep.
Use when a UI/UX project must verify that the promised user journey, TASK-PLAN.md, progress-state.json, rendered progress HTML, evidence artifacts, and skipped stages are consistent. Detects lazy skips, fake Ran/Done claims, missing evidence, approval-gated external steps, and user-data blockers; recommends continue, pause_for_user, return_to_rework, skip_with_reason, or block without replacing senior-ui-ux-orchestrator as chair.
In an external XGENT.ai Portal app's own repo, generate its companion CLAUDE.md / PRODUCT.md / DESIGN.md (with AGENTS.md written as a mirror) — read app.manifest.json and code facts, ask once about anything missing, fill in three ready-to-use documents from templates, leave no placeholders, and never overwrite existing files. Use when the task is "initialize/onboard a new external app repo", "add CLAUDE.md / AGENTS.md / PRODUCT.md / DESIGN.md to this app repo", "add a design document that impeccable can read", or when the user has just received an empty app repo or one with only code and no specs; apps inside the portal monorepo and non-XGENT projects do not use this skill. Use in an external XGENT portal app's own repo to scaffold its CLAUDE.md / PRODUCT.md / DESIGN.md (AGENTS.md is written as a mirror of CLAUDE.md) from the app manifest and repo facts — one round of questions, no leftover placeholders, never overwrites existing files.
Checklist and automation guide for adding a new skill to the OPC Skills project. Ensures all required files, metadata, logos, and listings are created before release. Use when adding a new skill, publishing a skill, or preparing a skill for release.
Build MCP servers in AdonisJS applications with the @jrmc/adonis-mcp package. Use when creating, editing, or debugging MCP tools, resources, or prompts in an AdonisJS app, when exposing AdonisJS functionality to AI assistants via the Model Context Protocol, or when setting up, securing, or testing an @jrmc/adonis-mcp server.
**WORKFLOW SKILL** — Create, update, review, fix, or debug VS Code agent customization files (.instructions.md, .prompt.md, .agent.md, SKILL.md, copilot-instructions.md, AGENTS.md). USE FOR: saving coding preferences; troubleshooting why instructions/skills/agents are ignored or not invoked; configuring applyTo patterns; defining tool restrictions; creating custom agent modes or specialized workflows; packaging domain knowledge; fixing YAML frontmatter syntax. DO NOT USE FOR: general coding questions (use default agent); runtime debugging or error diagnosis; MCP server configuration (use MCP docs directly); VS Code extension development. INVOKES: file system tools (read/write customization files), ask-questions tool (interview user for requirements), subagents for codebase exploration. FOR SINGLE OPERATIONS: For quick YAML frontmatter fixes or creating a single file from a known pattern, edit the file directly — no skill needed.
Audit, score, and compare repositories containing portable Agent Plugins against the official Agent Plugins specification. Use when asked to review a plugin repo, check plugin.json or mcp.json conformance, assess bundled skills and MCP servers, produce an evidence-cited 0–100 plugin scorecard, identify release blockers, or compare two agent plugins side by side.
Use this skill when users want to execute this workflow when a user wants to test how a different LLM model would handle existing Agentforce agent conversations. Trigger when users mention agentforce, offline, experimentation
Discover, select, and safely install focused skills from abagames/agentic-gamedev-skills. Use when a user explicitly wants that upstream collection, needs to choose among its mini-game design, Godot, crisp-game-lib, presentation, telemetry, or agent-workflow skills, or wants a pinned selective bundle install. Triggers on: agentic-gamedev-skills, abagames skills, install gamedev skill bundle, designing-mini-games, or upstream game skill inventory.
Wraps existing Python agent as Agent Stack service using agentstack-sdk with minimal compatibility changes and no business-logic rewrites. Use when migrating, wrapping, or deploying an existing plain Python or framework-based agent to Agent Stack; not for non-Python runtimes or building new agent from scratch.
Design co-learning experiences using the Three Roles Framework (AI as Teacher/Student/Co-Worker). Use when teaching AI-driven development workflows, spec-first collaboration, or balancing AI assistance with foundational learning. NOT for curriculum without AI integration.
Use for anything AI/agent/conversational built on Algolia: algobot CLI, Agent Studio, RAG systems, conversational product discovery, genAI content generation from search results (carousels, descriptions, headers), chatbots or recommendation agents using Algolia as retrieval, config-as-code workflows, multi-environment deploy (dev/staging/prod), memory and personalization, MCP tool integrations, conversation history / GDPR retention, or adding a chat widget alongside InstantSearch. Trigger on: "algobot", "Agent Studio", "RAG with Algolia", "conversational experience", "AI agent" + Algolia, "genAI carousel", "chat widget", or building AI features on top of Algolia search. Do NOT use for Agent Studio evaluations or eval-driven agent improvement - use agent-studio-evaluations. Do NOT use for raw index ops (records, synonyms, settings) - use algolia-cli. Do NOT use for pure frontend search UI (InstantSearch, autocomplete) with no AI/agent layer.
Gives the agent a cryptographic Alien Agent ID linked to a verified human owner, signed authentication to Alien-aware services over RFC 9449 DPoP, an encrypted vault for service credentials (GitHub, Slack, AWS, etc.), and SSH-signed git commits with provenance trailers. Use when the user asks to bootstrap an Alien identity, sign in to an Alien Network service (anything under alien-api.com / alien.org / agent-sso.*), post to or call such a service, store or retrieve service credentials securely, or make commits attributable to a specific agent + human owner. Also triggers on the terms "Alien ID", "Agent ID", "DPoP", "cnf.jkt", or "owner binding".
Records research provenance at the end of a coding or research session by scanning the conversation and writing decisions, experiments, dead ends, pivots, claims, and heuristics into the ara/ directory (exploration_tree.yaml, claims.md, heuristics.md, observations.yaml, session records), each tagged user, ai-suggested, ai-executed, or user-revised. Use when a task is finished and you want a session epilogue. Use when you need an auditable trace of how a project evolved. Use when decisions and dead ends need logging with provenance. Use when initializing an ara/ directory. Do not use during task execution, since it must run only after the request is complete.
Ask Grok or Gemini via opencli (browser-session based, no API cost). Use when the user says "问 grok", "问 gemini", "ask grok", "ask gemini", "用 grok 查", "grok 怎么看", "让 gemini 分析", or wants a second opinion from Grok/Gemini without paying API tokens. Routes to opencli which drives the already-logged-in Chrome session.
Orchestrates end-to-end autonomous AI research projects using a two-loop architecture. The inner loop runs rapid experiment iterations with clear optimization targets. The outer loop synthesizes results, identifies patterns, and steers research direction. Routes to domain-specific skills for execution, supports continuous agent operation via Claude Code /loop and OpenClaw heartbeat, and produces research presentations and papers. Use when starting a research project, running autonomous experiments, or managing a multi-hypothesis research effort.
Analyze recent conversation context and capture learnings to project knowledge files (for project-specific insights) or skills/commands/subagents (for cross-project patterns). Use when the user asks to "capture this learning", "update the docs with this", "remember this for next time", "document this issue", "add this to CLAUDE.md", "save this knowledge", or "update project knowledge". Also triggers after resolving build/setup issues, discovering non-obvious patterns, or completing debugging sessions with valuable insights.
Recall ChatCrystal memories for debugging tasks involving failing tests, compiler errors, runtime exceptions, dependency issues, environment breakage, or performance regressions. Use when historical root causes, fixes, or pitfalls may accelerate diagnosis before proposing a fix.
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.