Skills
101,323 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Initialize evo for the current repository by exploring the codebase, proposing unexplored optimization dimensions, constructing the benchmark inside a baseline worktree, and running the first experiment. Use when the user invokes /evo:discover, mentions setting up evo, wants to instrument a codebase for autonomous optimization, or asks to start a new evo run on a project.
Use this skill for hands-on DOCA Device Emulation on a BlueField DPU—exposing a custom emulated PCIe device that the host sees as a real peripheral while DPU-side code runs the backend; choosing the sub-library (PCI Generic, virtio-net, virtio-fs); wiring the per-sub-library Core context plus doorbell/DMA primitives; querying `doca_devemu_*_cap_*`; or debugging DOCA_ERROR_* from a `doca_devemu_*` call. Trigger even when the user doesn’t say “devemu”—typical implicit phrasings include “expose a custom PCIe device from BlueField to the host,” “host should see a virtio NIC backed by my DPU code,” “lspci does not show my emulated device,” “device enumerated but no driver binds,” “DPU sees nothing when host kicks the queue,” or “virtio feature negotiation failed at bind.” Refuse and route elsewhere for the packaged DOCA SNAP/Virtio-net Services, host-side virtio kernel drivers, backend body design, or standard BlueField NIC behavior—those belong to other skills.
ISV App Analytics metadata types — AppAnalyticsQueryRequest and AppAnalyticsSettings. Use this skill when the user asks about retrieving managed package usage data, configuring App Analytics simulation mode, querying subscriber snapshots, or understanding the AppAnalyticsQueryRequest lifecycle (New → Pending → Complete → Expired). TRIGGER when: user mentions App Analytics, AppAnalyticsQueryRequest, AppAnalyticsSettings, package usage data, subscriber analytics, ISV analytics, or simulation mode for app analytics. DO NOT TRIGGER when: the task is about standard Salesforce reports/dashboards (use reporting skills), custom SOQL on Account/Contact (use querying-soql), or Data Cloud query/search (use retrieving-datacloud).
Reference for fandhe-db-engine / fandhe-db-wire-server, the Rust-built local-first, vector-focused query database fandhe-db (compatible with PostgreSQL wire protocol v3). Covers USING PLAN / USING OPERATION_ID syntax, DDL / JOIN / CTE / window functions, extended queries, COPY and transactions, TLS 1.3 / SCRAM-SHA-256, the NoSQL HTTP layer, precision and recall modes, HNSW / BM25 / RRF hybrid / rerank, redb persistence, tenant / RLS, SIMD / wgpu, and the wire_code error contract.
Provides comprehensive guidance for FastAPI framework including routing, request validation, dependency injection, async operations, OpenAPI documentation, and database integration. Use when the user asks about FastAPI, needs to create FastAPI applications, implement REST APIs, or build high-performance Python web services.
Complete FFmpeg noise reduction and denoising for video and audio. PROACTIVELY activate for: (1) Video denoising (nlmeans, hqdn3d, vaguedenoiser), (2) Hardware-accelerated denoising (nlmeans_opencl, nlmeans_vulkan), (3) Audio noise reduction (afftdn, anlmdn), (4) Film grain removal, (5) Low-light footage enhancement, (6) Compression artifact removal, (7) Broadcast noise reduction, (8) Clean audio recordings. Provides: Denoising filters, parameter tuning, hardware acceleration, quality/speed tradeoffs.
Push design tokens from code INTO Figma as variables — DTCG / tokens.json / a token object → Figma variable collections, modes, and values. Use when the user wants to sync tokens code→Figma: triggers 'import tokens into Figma', 'create Figma variables from my tokens.json / DTCG / Tailwind config', 'sync design tokens to Figma', 'push tokens to Figma'. Non-destructively matches existing variables (by saved id/key/name) so re-imports update instead of duplicating. Works on ANY Figma plan. For the reverse (Figma → code) use figma-export-tokens.
Create, update, rename, delete, and organize Figma variables, collections, and modes — including fast batch create/update, setting scopes, code syntax, and adding theme modes. Use for incremental variable work: triggers 'create a variable', 'batch update variables', 'add a dark mode to my collection', 'rename this variable', 'set variable scopes', 'delete a variable/collection', 'add a mode'. Works on ANY Figma plan. For a from-scratch system use figma-setup-design-tokens; for code↔Figma sync use figma-import-tokens / figma-export-tokens.
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Fix knip "Unused exports" violations. Handles all violation categories: test-only exports (extract to new file), dead barrel re-exports (remove from index.ts), and internally-only-used exports (un-export). Use when `npm run knip` reports unused exports.
Proactive "make the most of gbrain" coaching. Runs `gbrain advisor` on a cadence and pings the user with the top high-leverage actions for their brain: version drift, pending migrations, stalled jobs, low embed coverage, setup smells, and uninstalled brain skills. Read-only; always asks before fixing.
Build explorative, interactive learning experiences as Next.js apps using the Geist design system. Use when creating tutorials, explorable explanations, interactive lessons, code sandboxes, quizzes, or any educational UI. Covers the Learning Loop pedagogy, 23+ learning component patterns, progress tracking, spaced repetition, and Bret-Victor-style interactive exploration — all with Geist's dark-first minimal aesthetic.
Execute git commit with conventional commit message analysis, intelligent staging, and message generation. Use when user asks to commit changes, create a git commit, or mentions "/commit". Supports: (1) Auto-detecting type and scope from changes, (2) Generating conventional commit messages from diff, (3) Interactive commit with optional type/scope/description overrides, (4) Intelligent file staging for logical grouping
Build Google Ads campaign structures with keyword research, ad copy, audience targeting, and budget allocation. Use when planning paid search campaigns, setting up Google Ads for the first time, or restructuring underperforming ad accounts. Supports B2B and B2C.
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.
Universal AI voice / text-to-speech skill supporting OpenAI TTS (gpt-4o-mini-tts, tts-1), ElevenLabs multilingual TTS with voice cloning, Bailian Qwen TTS (qwen-tts / qwen3-tts-vd with voice-design custom voices, long-text chunking built in), MiniMax speech-02-hd, SiliconFlow CosyVoice / SenseVoice, and PlayHT 2.0. Use this skill whenever the user asks to read text aloud, synthesize speech, generate narration, create voice-over, dub a script, or turn any text into audio (mp3 / wav / ogg / flac). Typical phrases include "read this aloud", "generate voice for ...", "create a narration of ...", "tts this", "把这段念出来", "做个配音", "合成语音", or mentions of voices / TTS model names like Alloy, Ash, Cherry, Rachel, CosyVoice, PlayHT. Always use this skill even if the user does not specify a provider — pick one from EXTEND.md defaults or available env keys
Purchase a Huawei Cloud X instance server and one-click deploy the SQLBot intelligent query application. Tech stack: Python 3.8+, Huawei Cloud SDK, COC (Cloud Operations Center) deployment. Key capabilities: AK/SK authentication, X instance creation, automatic security group configuration, and automatic SQLBot deployment. Use cases: Users who need to quickly deploy the SQLBot intelligent query application on Huawei Cloud X instance servers. Trigger words: “deploy sqlbot”, “install sqlbot”, “sqlbot deploy”, “one-click deploy sqlbot”, “x-instance deploy sqlbot”, “部署sqlbot”, “安装sqlbot”, “sqlbot部署”, “一键部署sqlbot”, “x实例部署sqlbot”.
Edit, review, audit, or draft standalone prose while preserving the writer's meaning and voice. Use for documentation, markdown, emails, blog posts, PRDs, and other dedicated writing when the user wants clearer, more direct, more natural, or less AI-patterned language, or asks whether a draft contains AI-slop patterns. Do not use for routine conversational replies or code unless the user asks to edit its prose.
Use when pausing, shutting down, handing off, or context-switching active work and future sessions need a compact resumable checkpoint derived from the current session context. Writes a small checkpoint note under wiki/checkpoints/ and then optionally records the user's intended return step. Not for durable learnings capture, wiki correction, or source ingestion.
Use when an approved workspace spec needs an execution-ready implementation plan with ordered, verifiable steps. Specs are mandatory: this skill consumes design decisions, verifies that the spec still matches the codebase, and writes the repo-native plans/ artifacts. Inherits optional goal provenance; see loam::setting-goals.
Create or extend an Obsidian-friendly markdown wiki scaffold in the current workspace. Use this when the user wants to build a wiki, set up a knowledge base, create a research vault, or scaffold a living markdown note graph before adding sources with `/loam::adding-to-memory`. Identifies goals as sibling workflow artifacts, not wiki content. Not for importing or normalizing existing memory-like corpus; use /loam::normalizing-memory for that.
Polish and elevate MBA thesis / dissertation to national outstanding thesis quality (全国优秀论文). Performs comprehensive improvement: academic language, argument structure, logical rigor, innovation highlights, and formatting. Input: markdown thesis text. Output: polished full text. Also trigger when the user mentions "论文润色", "MBA论文", "优秀论文", "thesis polish", "dissertation improvement", "学术润色".
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.
Enforce and guide the Mikado Method for large, graph-driven refactors where a single naive change cascades into many failing call sites. The skill enforces the full Mikado loop: goal → naive attempt → map prerequisites → revert → implement leaves → commit → repeat. TRIGGER when: user mentions Mikado Method, dependency untangling, circular dependencies, "breaking everything" on a refactor, changes cascading across many files, ripple effects, large-scale legacy refactoring with tight coupling, untangling tightly coupled modules, splitting a big refactor into safe steps, or wanting to work on main without a long-lived feature branch. Also trigger on "graph-based refactor", "prerequisite tree", or "Mikado graph". DO NOT USE when: the change is small/in-place (single method, single class, simple rename) — use `clean-code/refactoring` instead for code smells and Extract Method-style work. If the feature hasn't been scoped yet, consider `kano` first to validate it's worth doing.
ALWAYS load when writing or debugging code that imports "motion-v" (Motion for Vue, motion-vue). Covers the motion component, AnimatePresence, variants, gestures, layout, scroll, motion values, the v-motion directive, and the Nuxt module for version 2.4.4. Provides API facts, recipes, best practices, and migration notes
A complete guide to developing MusicFree desktop theme packs from scratch. Triggered when the user asks to write, create or design a MusicFree desktop theme pack, or to generate a theme from a reference image, colors or style keywords. Covers the full workflow: the CSS variable system, color design paradigms, static themes, dynamic iframe themes, asset optimization, packaging and testing, and submitting to the theme marketplace. This Skill is designed for AI execution, guiding AI and community contributors (who may have no frontend experience) to collaborate on theme pack development.
Reference for ngrok's traffic model—what an endpoint is and what happens to traffic at the edge. Covers endpoint types (agent, cloud, internal), bindings and protocols (HTTP, TLS, TCP), and Traffic Policy—the YAML rule language of phases, CEL expressions, and actions—with a generated catalog of all 26 actions. Other ngrok skills refer here to decide what to build, then use ngrok-surfaces to apply it. Not a standalone task; load when another skill references it.
Huawei Ascend NPU npu-smi command reference. Use for device queries (health, temperature, power, memory, processes, ECC), configuration (thresholds, modes, fan), firmware upgrades (MCU, bootloader, VRD), virtualization (vNPU), and certificate management.
Run Comet's Opik — open-source LLM observability, evaluation, and optimization — from one routing-first skill: install the Python/TypeScript SDK, stand up a server (Comet.com cloud, Docker Compose via `./opik.sh`, or Kubernetes/Helm), wire tracing through `@opik.track` or one of 50+ framework integrations (OpenAI, Anthropic, LangChain, LangGraph, LlamaIndex, CrewAI, DSPy, Ollama, Bedrock, Vercel AI SDK, …), score outputs with LLM-as-a-judge metrics (Hallucination, Moderation, Answer Relevance, Context Precision), and run Datasets/Experiments evaluations including PyTest CI gates. Use when the user wants LLM tracing, prompt evaluation, production LLM monitoring, agent optimization, or guardrails with Opik. Triggers on: opik, comet opik, opik configure, opik.sh, llm observability, llm tracing, llm as a judge, hallucination metric, prompt evaluation, opik dashboard, opik guardrails, agent optimizer.
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.
Expert partnership marketing guidance for building and scaling partner programs, co-marketing campaigns, and channel ecosystems. Use when developing partner strategy, creating co-marketing content, designing partner tiers, building affiliate programs, positioning in integration marketplaces, enabling channel partners, planning joint webinars, or structuring revenue sharing. Use for technology partnerships, reseller programs, strategic alliances, and partner relationship management.
Helps diagnose underperformance, structure a performance case, decide whether to put someone on a PIP or let them go, and think through your own manager review. Use when the user says "perf review," "review cycle," "performance calibration," "write a review," "rating," "meets expectations," "performance improvement," "PIP," "promotion case," "end of year review," "mid-year review," "underperformer," "struggling employee," or "should I let them go." Do NOT use for delivering day-to-day feedback (use feedback) or for evaluating a new hire candidate (use hiring).
Pillar smart wallet operations in two modes—browser-handoff (pillar.ts) opens the Pillar frontend for user signing, and agent-signed direct (pillar-direct.ts) signs locally with a secp256k1 keypair and submits directly to the Pillar API (no browser required, gas sponsored). Supports sBTC send/supply/boost/unwind, DCA programs, stacking, key management, wallet creation, and position queries.
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.