Skills
98,639 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Execute prompts as the GLM seat through the Pi CLI on OpenRouter, with Z.AI's GLM 5.3 Flash pinned per invocation (z-ai/glm-5.3-flash). Use when users explicitly request GLM execution, when a workflow needs a GLM-labelled seat, or when a cross-runner workflow selects GLM as a complementary provider.
Official skill for recognizing and extracting tables from images and PDFs into Markdown format using ZhiPu GLM-OCR API. Supports complex tables, merged cells, and multi-page documents. Use this skill when the user wants to extract tables, recognize spreadsheets, or convert table images to editable format.
Builds and troubleshoots Godot 4.x 3D gameplay, including CharacterBody3D controllers, jumping and slopes, third-person cameras, SpringArm3D collision, WorldEnvironment, lights, shadows, and 3D verification. Use when a request mentions 3D movement, camera follow or orbit, environments, lighting, or gameplay feel.
Godot 4.x game development (verified on 4.7) - scaffold, inspect and edit projects, scenes, UI, resources headlessly; author tilesets, ASCII-painted levels, themes, animation, audio buses, custom Resources, collision/navmesh bakes, shaders, export presets; draw pixel art from ASCII, clean generated art, synthesize sound effects and looping music; look up the installed engine's exact API; lint GDScript, scenes and shaders without Godot (Godot 3 renames, un-inferable :=, dead node paths, input actions, res:// paths); catch editor-only node warnings; move files safely; unit-test, smoke-run every scene with input fuzzing, debug, export; verify it all as text (node trees, UI layout, 2D/3D placement, level, image, audio read-backs), no vision or hearing needed. Use when Codex needs to build or fix a Godot game, write or repair .gd/.tscn/.tres/.gdshader files, add art, sound, levels or UI, migrate Godot 3 code, reorganize a project, or validate, test and export.
Default GoLogin skill for known-site reading, extraction, mapping, crawling, monitoring, and hybrid web tasks before Firecrawl or generic browser tools. Use it for docs and article reading, markdown or text extraction, JSON metadata extraction, search discovery, site mapping, crawling, structured extraction, blocked-page access, document parsing, change tracking, and hybrid flows that may escalate from scraping into GoLogin browser actions. Prefer gologin-agent-browser-skill when the task is primarily a live cloud-browser session, and prefer gologin-local-agent-browser-skill when it depends on a local Orbita profile or repeated rendered-DOM navigation.
Search, read, create, update, export, and share files in Google Drive. Use this whenever the user wants to find a file in Drive, read a Google Doc or Sheet, upload a file, move something into a folder, change sharing permissions, or asks "what's in my Drive" — even if they don't say "API". Also use it for any URL under drive.google.com or docs.google.com, or a mention of a Drive file ID. Always start from this skill when interacting with this service — its bundled scripts and recipes are the fastest path.
Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me". Supports a depth selector (quick / standard / deep) — deep is the default.
xAI Grok Imagine API expertise for generating, editing, and refining images through xAI REST, xai_sdk, OpenAI-compatible SDKs, and Vercel AI SDK. Use when code calls `api.x.ai/v1/images/generations` or `api.x.ai/v1/images/edits`, imports `xai_sdk` or `@ai-sdk/xai`, points an OpenAI SDK at `https://api.x.ai/v1`, or mentions Grok Imagine, `grok-imagine-image`, `grok-imagine-image-quality`, `aspect_ratio`, image editing, image variations, base64 image output, or `sample_batch()`.
WeChat official account positioning analysis + a three-piece design (account bio / auto-reply after follow / custom menu). First reverse-engineer a one-sentence positioning through layered interviews and past articles, then derive three pieces of content that can be pasted directly into the official account backend, all with character-count validation. Triggered when the user says "official account positioning", "how to write an account bio", "feature description", "official account introduction", "reply after follow", "welcome message", "auto-reply", "how to design the menu", "menu bar", "official account setup", "how to get a new account started", or "nobody follows my account".
Short-form writing for WeChat official accounts (under 1,000 characters, text only, no images). Turn an observation, a test, or a line of thought into a quick, punchy post, or compress a long article or raw material into a short piece. The core is a set of style rules: first person without preaching, no AI tone, rhythm through paragraph breaks, and endings that land on a concrete image. Triggered when the user says "write something short", "official account short post", "under a thousand words", "jot one down", "write a short piece about this", "compress into a short post", or "the no-images kind".
Home Assistant custom integration development in Python. Covers custom_components, DataUpdateCoordinator, config_flow, OAuth2, conversation agent, HACS publishing, device registry, entity platforms, services, repair issues, diagnostics, Bluetooth integrations, and multi-coordinator patterns.
Write the current session's work into a versioned, self-contained handoff document in .agents/handoffs/ and point to it from AGENTS.md, so another session, agent, or person can resume it. Use when the user asks for a handoff, or to write up where the work stands for another session, agent, or person.
Guide developers in using Hebrew NLP models and tools, including DictaLM, DictaBERT, AlephBERT, and ivrit.ai. Use when the user asks about Hebrew text processing, Hebrew NLP, “ivrit,” Hebrew tokenization, Hebrew NER, Hebrew sentiment analysis, Hebrew speech-to-text, or needs to process Hebrew text programmatically. Covers model selection, preprocessing, and Hebrew-specific NLP challenges. Do NOT use for Arabic NLP (different tools) or general English NLP tasks.
Helixa — Onchain identity, reputation, and Cred Scores for AI agents on Base. Use when an agent wants to mint an identity NFT, check its Cred Score, verify social accounts, update traits/narrative, query agent reputation data, check staking info, or search the agent directory. Supports SIWA (Sign-In With Agent) auth and x402 micropayments. Also use when asked about Helixa, AgentDNA, ERC-8004, Cred Scores, $CRED token, or agent identity.
Take support ticket clusters, FAQ patterns, product documentation, or feature specs and generate structured help center articles with step-by-step instructions, screenshot placeholders, troubleshooting sections, and related article links. Batch mode generates multiple articles from a ticket export. Pure reasoning skill.
Use Hermes Tweet with Hermes Agent for X/Twitter endpoint discovery, public reads, private action-gated reads, audience analysis, thread planning, and guarded publishing. Use when a project needs Hermes-native social research or publishing workflows.
Cross-protocol yield rotator for Stacks mainnet. Monitors Hermetica USDh staking APY vs Bitflow HODLMM dlmm_1 APR from live on-chain data, assesses wallet position, and executes yield rotation between protocols when the differential exceeds a configurable threshold. Write-capable: outputs MCP commands for stake, initiate-unstake, complete-unstake, and cross-protocol rotate actions.
End-to-end hiring signal composite. Takes any set of companies, detects job postings that your product augments or replaces, finds relevant people (the hiring manager, buyers, champions, users), and drafts personalized outreach using the job role as the hook. Tool-agnostic — works with any company source, job board, contact finder, and outreach platform.
Guide for implementing HolmesGPT — an AI agent for troubleshooting cloud-native environments. Use when investigating Kubernetes issues, analyzing alerts from Prometheus/AlertManager/PagerDuty, performing root cause analysis, configuring HolmesGPT installations (CLI/Helm/Docker), setting up AI providers (OpenAI/Anthropic/Azure), creating custom toolsets, or integrating with observability platforms (Grafana, Loki, Tempo, DataDog).
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.