Skill rating

Build a real, editable PowerPoint (.pptx) deck from an outline or brief. Use when asked to make a slide deck, a PowerPoint, a pitch/board/sales deck as an actual file, or to turn a doc/notes into slid

Compress tool outputs, logs, and JSON before they enter the context window — structural compression via a deterministic stdlib script (schema + samples + stats instead of 300 raw rows), no API, no sum

Navigate a codebase by map instead of reading files wholesale — a deterministic stdlib script that emits the tree with line counts and top-level symbols, plus the read-the-map-first discipline that cu

Screen and evaluate resumes against criteria using ZhiPu GLM-V multimodal model. Reads multiple resume files (PDF/DOCX/TXT), compares against user-defined screening criteria, and outputs a Markdown ta

Analyze images/videos and generate professional prompts for text-to-image and text-to-video AI tools (Midjourney, Stable Diffusion, DALL-E, Sora, Runway, Kling, Pika). Use when the user wants to gener

Trigger when: (1) User wants to extract text, tables, formulas, or structured data from images/PDFs/scanned documents, (2) User mentions "OCR", "文字识别", "文档解析", (3) User has a document (screenshot, sca

Design, implement, harden, and verify Model Context Protocol (MCP) servers with precise tool contracts, least-privilege authorization, safe transports, structured errors, and interoperability tests. U

Design and verify auditable human oversight, approval gates, escalation paths, and safe state transitions for AI agent workflows. Use when deciding which agent actions require review, adding approve/r

Audits and improves how a website is found and cited by AI answer engines (Google AI Overviews and AI Mode, ChatGPT search, Perplexity, Claude, Copilot). Use for Generative Engine Optimization (GEO, A

Installs a Claude Code PreToolUse hook that stops destructive git commands before the shell runs them: force pushes, pushes to protected branches, reset --hard, clean -f, checkouts and restores that d

Plans, researches and writes comparison pages: "[rival] alternatives", "[product] vs [rival]", "[rival A] vs [rival B]" and switching guides. Builds each page from a ledger of sourced, dated facts and

Applies behavioural science to marketing decisions using only effects with solid published evidence, named as the research literature names them, with their replication record. Use when someone asks a

Rebuilds how a past feature, fix or release came about by reading the local ax history graph (sessions, commits, subagents) through the ax CLI. Use when asked "how did we build X", "what made Y work",

Audits and reorganises what an AI coding agent carries in its context window: instruction files, path-scoped rules, skills, subagents, hooks and the running conversation. Use when someone says "the ag

Designs and audits the moments inside a product where an account is asked to pay or to move to a higher plan: feature gates, usage-limit screens, trial-ending and trial-expired states, seat and add-on

Audits a repository to decide which agent skills it should have: inventories the skills and instruction files already present, mines session history, memory, scripts and commits for procedures that ke

Builds an iOS app, installs and launches it on a simulator, then collects evidence about a runtime problem: unified logs, console output, screenshots, the accessibility tree, crash reports and an LLDB

Cuts LLM API spend by measuring cost per feature and user, routing easy requests to cheaper models, using provider prompt caching, adding a semantic response cache, and using the batch APIs. Use when

Query the Unusual Whales API for unusual options flow, dark pool prints, market tide sentiment, gamma exposure, congressional trading, and stock greeks. Use when someone asks about "unusual options ac

Explains how to find, choose, install and invoke agent skills (SKILL.md folders that follow the Agent Skills specification) in Claude Code, Codex, Gemini CLI and Cursor, each by its own documented mec

When the user wants to set up authentication and authorization for a web application. Use when the user mentions "auth," "login," "OAuth," "SSO," "single sign-on," "role-based access," "RBAC," "permis

When the user needs to design database schemas and queries optimized for real-time applications. Use when the user mentions "chat database," "message storage," "real-time sync," "message history," "un

Google AI Studio and the Gemini API give programmatic access to Gemini models for multimodal (text, image, video, audio, PDF) generation, long context, structured JSON output, function calling and Goo

Write specifications before code — the agent creates detailed technical specs, validates them, then implements. Use when someone asks to "write a spec first", "plan before coding", "create a technical

Discover and install AI agent skills from terminalskills.io. Use when the user asks for a capability the agent does not already have a skill for — search the catalog by keyword, category, or tag; insp

Audits and redesigns what happens between account creation and a new user's first real result, so that more signups activate and come back. Use when someone says "activation rate is low", "users sign

Aceternity UI is a collection of animated React components (3D cards, spotlight, background beams, moving borders, wavy backgrounds) built with Tailwind CSS and the motion library, installed into your

Run AI agent code safely in isolated sandboxes with resource limits, audit trails, and kill switches. Use when someone asks to "sandbox my agent", "run agent code safely", "add guardrails to AI agent"

Extracts text and its structure from an image by the route that fits the picture: an OCR engine for clean printed text, the agent's own vision for photos, tables, forms and handwriting, or both, and t

Add persistent memory to AI coding agents — file-based, vector, and semantic search memory systems that survive between sessions. Use when a user asks to "remember this", "add memory to my agent", "pe

The AI Scientist-v2 is Sakana AI's open-source research agent that turns a topic description into research ideas, runs machine-learning experiments with an agentic tree search, and writes a LaTeX pape

Reviews a diff with several independent read-only reviewer agents running at the same time, each looking through one lens (correctness, security, data, concurrency, compatibility, tests), then checks

Browserbase runs managed headless Chromium sessions in the cloud, so browser automation, AI agents and scraping jobs need no browser infrastructure of their own. Use when someone asks to "run Playwrig

Hugging Face is the hub and Python libraries (transformers, datasets, peft, huggingface_hub) for downloading, running, fine-tuning and publishing machine-learning models. Use when the user asks to "lo

Tailwind CSS is a utility-first CSS framework: you style pages with small classes in the markup and ship only the CSS you use. Covers Tailwind v4 (CSS-first config with @theme, @import "tailwindcss",

Run AI agent and LLM evaluations in CI/CD pipelines — automated quality gates that fail the build when AI output quality drops. Use when someone asks to "test my AI agent", "add evals to CI", "catch p

Use Tailwind CSS in React Native with NativeWind — write className instead of StyleSheet. Use when someone asks to "use Tailwind in React Native", "NativeWind", "style React Native with Tailwind", "cl

Connects a coding agent to Google Stitch through the stitch-mcp CLI and MCP proxy, so the agent can read a screen's HTML and screenshot and rebuild it in the project's own framework. Use when a user s

Specifies, builds and audits website popups: modal dialogs, slide-ins, sticky bars and exit prompts that collect an email, show an offer or make an announcement. Picks the format and trigger, keeps th

Stagehand is Browserbase's open-source SDK for browser agents. It adds three natural-language calls (act, extract, observe) on top of a Chrome browser you drive from TypeScript, Python or Go. Use this

Traceloop's OpenLLMetry SDK adds OpenTelemetry tracing to LLM applications: it auto-instruments OpenAI, Anthropic, LangChain, LlamaIndex and vector databases, and exports spans to any OTLP backend (Gr

Composio connects AI agents to 1,000+ external apps (GitHub, Slack, Gmail, Notion, Linear and more) and handles OAuth, token storage and per-user connections for them. Use when someone asks to "give m

Guidance is Microsoft's open-source Python library for controlling LLM output with constrained generation: you interleave text, control flow, regex, select() choices and JSON schemas, and the model ca

Helicone is an LLM observability platform and proxy that sits between your app and providers such as OpenAI and Anthropic, logging every request with cost, latency and user metadata, and adding cachin

NemoClaw is NVIDIA's open-source reference stack that installs and runs an AI agent (OpenClaw by default; Hermes and LangChain Deep Agents Code also supported) inside an NVIDIA OpenShell sandbox with

pgvector is a PostgreSQL extension that stores vector embeddings in ordinary tables and searches them by similarity, so no separate vector database is needed. Use when someone asks for "vector search

LanceDB is an embedded, serverless vector database that stores data in the Lance columnar format on local disk or object storage, with Node.js, Python and Rust clients. Use when someone asks for "vect

The Vercel AI SDK is a TypeScript toolkit for calling LLMs from Node.js, Next.js, Nuxt, SvelteKit and edge runtimes through one API. It covers text generation, streaming, structured output with Zod, t

Run Python functions on cloud GPUs and CPUs with Modal, a serverless platform that builds containers from Python code, scales to zero and bills per second. Use when a user asks to "run this on a GPU",

Xano is a no-code/low-code backend platform for building REST APIs, a PostgreSQL database and user authentication without server code. Use when someone asks to "build a backend with Xano", "create a X