Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Testing & quality
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs
Use when an issue needs deeper analysis than triage — confirming root cause by reading code, verifying file/line references in the body are still valid, checking blast radius across the codebase, or proposing a concrete fix direction. One issue per invocation.
Expert debugging and profiling for leaks, GPU/Visual Profiler, headless CI QA, orphan nodes, thread-safe logs, and custom Debugger monitors — not print/breakpoint tutorials. Trigger on OBJECT_ORPHAN_NODE_COUNT, ObjectDB growth, Visual Profiler GPU spikes, flaky headless exits, or remote device consoles. Keywords: orphan nodes, Performance.get_monitor, Visual Profiler, Time.get_ticks_usec, EditorDebuggerPlugin, headless QA, push_error, backtrace.
Use SurrealKit, SurrealDB's schema-management and migration CLI, to scaffold projects from templates, sync schema in development, plan and execute production rollouts (with rollback), statically check SurrealQL without a database, generate JSON/TypeScript types, and write declarative TOML test suites for schemas, permissions, and API endpoints. Use this skill whenever users set up, migrate, type, check, or test a SurrealDB schema with SurrealKit.
Generate a project-local verify-<app> skill on top of e2e, so any coding agent can launch the app, check it is worth driving, exercise a feature the way a user does, keep evidence, and bug bash it. Writes the skill, a feature map with one explore charter per feature, and proves it once. Use for /create-verification-skill, "make a verification skill for this repo", or when a project has no scripted way to prove UI behavior.
Turn a real agent failure into one reproducible eval-harness case plus the smallest knowledge change — failing case first, one knowledge owner, semantic assertions, before/after evidence through the repo's configured harness commands. Use for "add harness case", "agent got this wrong", "extend the harness", "rozszerz harness".
Writes turn-level tests for a LiveKit agent in the user's normal test suite: pytest (Python) or Vitest (Node.js). Use when the user asks to "write tests for my agent", "add a test for this tool", "test the handoff", "pin this bug", "why does my agent test fail", or after building or changing agent behavior that needs regression coverage. Covers the SDK's test session harness, assertions on messages, tool calls and handoffs, LLM judging of a reply against an intent, mocking tools, multi-turn tests, and judging whole conversations with the built-in judges. For interactive poking use debugging-livekit-agents. For grading whole conversations at scale use running-livekit-simulations.
LLM-as-judge for a coding-agent session or an eval-harness result — checks controller attestations first, then project guards, code review, and the repo's design contract, and names the smallest harness owner to fix. Read-only; never executes session content. Use for "judge this session", "analyze this eval", "oceń sesję/agenta".
Perform comprehensive code reviews using OpenAI Codex CLI. This skill should be used when users request code reviews, want to analyze diffs/PRs, need security audits, performance analysis, or want automated code quality feedback. Supports reviewing staged changes, specific files, entire directories, or git diffs.
Investigate AI observability evaluations — `hog` (deterministic code-based), `llm_judge` (LLM-prompt-based), and `sentiment` (user-message sentiment). Find existing evaluations, inspect their configuration, run them against specific generations, query individual results, and set up scheduled reports on an evaluation. Use when the user asks to debug why an evaluation is failing, surface common failure modes, compare results across filters, dry-run a Hog evaluator, prototype a new LLM-judge prompt, inspect sentiment classifications, or manage the evaluation lifecycle.
Diagnose a failure from its symptom — reproduce, classify the broken invariant, route via the repo AGENTS.md Task Router, trace the smallest root cause, and on request land a minimal fix behind a regression test. Framework knowledge via om-framework-context. Use for "why does this fail", "debug this", "fix this bug", "regression", "napraw błąd".
Use when a developer wants to iterate on ONE specific Agent Observability / LLM Obs trace whose output they didn't like—rerunning it against local code, viewing a concise diff of the old and new output, and looping (change code → replay → diff) until satisfied. Invoked as /agent-observability-replay-trace <trace-id> [changes to test]. Signals: "replay this trace"; "iterate on a trace"; "this trace's output is wrong, fix it and re-run"; "re-run trace <id> with <change>"; or pasting a trace ID from the Agent Observability UI with a description of what to fix. It fetches the trace via the datadog-llmo MCP or the pup CLI, edits code, reruns the app to emit a new trace, and diffs the two—no local server or browser. For Agents traced with ddtrace / LLM Obs (Python first-class), with JSON-serializable entry input. Do NOT use for: scored Experiments or the browser "Replay" button (that's agent-observability-replay-experiment), building an experiment from a dataset/CSV, writing evaluators, root-causing failed traces, or RUM/HTTP session replay.
Debug and inspect LLM/AI agent traces using PostHog's MCP tools. Use when the user pastes a trace or session URL (e.g. /ai-observability/traces/<id> or /ai-observability/sessions/<id>), asks to debug a trace, figure out what went wrong, check if an agent used a tool correctly, verify context/files were surfaced, inspect subagent behavior, investigate LLM decisions, or analyze token usage and costs. Also use when raw SQL/HogQL against `events.properties.$ai_input` / `$ai_output_choices` returns empty — message content lives only on the dedicated `posthog.ai_events` table.
Draw what an app's user sees as a Mermaid statechart from a TLA+ model. States are gray if they work, red if still wrong, and green if fixed. Each bug TLC finds is a numbered path. Use for /statechart, \"show the PM the bug\", or to turn an agreed chart into a TLA+ skeleton.
Test-Driven Development methodology and red-green-refactor workflow (formerly test-tdd). Use this skill when practicing TDD, writing tests first, designing tests before implementation, or reviewing test-first approaches. Triggers on “write tests first,” “test before code,” “red green refactor,” “test driven development.” This skill does NOT cover Vitest framework specifics (use vitest skill) or API mocking with MSW (use msw skill).
Leverage OpenAI Codex/GPT models for autonomous code implementation, code review, and plan review. Triggers: "codex", "use gpt", "gpt-5", "let openai", "full-auto", "adversarial review", "second opinion review", "用codex", "让gpt实现", "对抗式审查", "让codex审查计划", "第二意见". Use this skill whenever the user wants to delegate coding tasks to OpenAI models, run code or plan reviews via codex, get a second-opinion review from a different model, or execute tasks in a sandboxed environment.
Collect and analyze on-device performance metrics and crash diagnostics using MetricKit. Use when setting up MXMetricManager, handling MXMetricPayload or MXDiagnosticPayload, processing crash/hang/disk-write diagnostics via MXCallStackTree, adding custom signpost metrics, or uploading telemetry to an analytics backend
Audit PostHog experiments and feature flags for configuration issues, staleness, and best-practice violations. Read when the user asks to audit, health-check, or review experiments or feature flags, check flag hygiene, or verify experiment setup.
Checks requirements.txt, pyproject.toml, and package.json dependency manifests for surface-level direct-dependency footguns: standard-library shadowing pins, abandoned backports, unpinned dependencies, and obvious intra-manifest conflicts, plus opt-in PyPI yanked releases. Use when the user asks to check a manifest for dependency problems, asks why dependencies won't install or whether anything is wrong with their dependencies, wants a dependency autopsy, or suspects dependency manifest rot. Runs offline by default as a local tool for the user's own project, not repository CI
Thorough, file-cited technical debt audit across 9 dimensions using AST-grep (tree-sitter), grep, LSP, and language-native tooling. Produces TECH_DEBT_AUDIT.md with severity, effort estimates, and prioritized fixes. Use when asked for codebase health check, tech debt audit, architecture review, code quality assessment, or cleanup planning. Triggers: 'tech debt', 'technical debt', 'debt audit', 'code health', 'technical debt audit', 'codebase health check', 'find tech debt', 'debt analysis', 'audit code quality'.
This skill should be used when the user asks to "debug this", "fix this error", "investigate this bug", "troubleshoot this issue", "find the problem", "something is broken", "this isn't working", "why is this failing", or reports errors/exceptions/bugs. Provides systematic debugging workflow and common error patterns.
Audit iOS and Android mobile applications against OWASP MASVS / MASTG — insecure storage, weak crypto, certificate pinning, deeplinks, IPC, jailbreak/root detection, reverse-engineering resistance. Use when the user mentions 'mobile security,' 'iOS security,' 'Android security,' 'mobile audit,' 'mobile pentest,' 'MASVS,' 'MASTG,' 'certificate pinning,' 'jailbreak detection,' 'root detection,' 'deeplink,' 'URL scheme,' 'app transport security,' 'keychain,' 'keystore,' 'mobile reverse engineering,' or has a mobile app to review.
Integrate Mem0 into an existing repository using a goal-driven, TDD pipeline. Automatically detects the repo's language and asks the user to choose between Mem0 Platform (managed) and Mem0 Open Source (self-hosted). Writes failing tests before implementation. Produces a local feature branch plus `.mem0-integration/` artifacts consumed by the paired verification skill. TRIGGER when: the user says "integrate mem0", "add mem0 to this repo", "wire mem0 into <repo>", or asks how to add memory to an existing project. DO NOT TRIGGER when: the user wants general SDK usage (use skill:mem0), CLI usage (use skill:mem0-cli), or Vercel AI SDK (use skill:mem0-vercel-ai-sdk). After success, invoke skill:mem0-test-integration to verify in the same workspace (loose coupling).
React Testing Library best practices for writing maintainable, user-centric tests. Use when writing, reviewing, or refactoring RTL tests. Triggers on test files, testing patterns, getBy/queryBy queries, userEvent, waitFor, and component testing.
Code-reviews Pluggy API integrations against Pluggy's official documentation (queried in real time via the Pluggy MCP, with a web fallback to docs.pluggy.ai when the MCP isn't connected) and returns a diagnostic report (✅/❌/⚠️) with file, line, and the code fix for each issue. Use WHENEVER the dev uploads integration files and asks to review, analyze, diagnose, or validate their Pluggy integration — or says things like "review my integration", "check my Pluggy code", "is this ready for production?", "Pluggy Doctor", "check my webhooks", "is my integration secure?", or the Portuguese variants "analisa minha integração", "revisa meu código da Pluggy", "tá tudo certo pra ir pra produção?", "checa meus webhooks", "minha integração tá segura?". Also trigger when the dev pastes/uploads code that clearly calls the Pluggy API (connect_token, GET /items, item/created webhooks, clientUserId, etc.) and wants to know if it's correct, even without saying "Pluggy Doctor". This skill is for REVIEWING existing code, not writing an integration from scratch.
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.