도구

Adds /learn commands to 30+ agents including Claude Code, Codex and Cursor: knowledge maps, layered Socratic explanations, quizzes and spaced review, with a visual learning dashboard. Geared to self-study of programming and other technical topics; needs Node.js 20+.

Automatically aligns subtitles to audio and video with a neural network, and can transcribe subtitles with Whisper or translate them with local models; supports 10+ formats such as SRT, TTML and SCC. Needs Python 3.9-3.12 and FFmpeg; transcription and translation need the llm extra.

Semantic search over past Claude Code, Codex and other agent sessions: indexes each conversation when it ends and recalls past discussions, decisions and trade-offs via MCP tools and a subagent. Embedding and search run locally. By the author of Superpowers.

Official Tavily skills: let agents search the web, extract page content, discover site URLs, crawl in bulk and run cited deep research through the Tavily CLI (tvly). Install the CLI first; search and extract have a limited no-login tier, other features need sign-in or an API key.

Lets agents work with Salesforce orgs authorized on your machine: SOQL queries, metadata deploy and retrieve, Apex test runs and permission set assignment, plus toolsets for LWC development. Maintained by Salesforce; authorize your org with the Salesforce CLI first.

Write a script, speakers and visuals in MulmoScript (JSON or YAML) and the mulmo command generates narration and images and assembles a video, podcast, slide deck or PDF, for explainer videos. Needs Node.js, FFmpeg and an OpenAI API key; other providers optional.

Astronomer's official data engineering plugin: write Airflow DAGs to best practices, debug pipeline failures, trace data lineage, profile warehouse tables and migrate from Airflow 2 to 3, with a built-in Airflow MCP server and session hooks.

Research guardrails for PhD students: reproduce arXiv papers, debug training anomalies with evidence, compare runs at matched epochs, check paper-code consistency, verify BibTeX, find literature gaps and predict reviewer comments. No MCP or extra dependencies; mostly for ML research.

Typed as /bro, has the agent re-explain its previous reply in plain, casual language without answering anything new or calling tools, keeping paths, commands and numbers verbatim and replying in the original language.

Anthropic's official skill that reads outpatient notes like a medical coder, decides which diagnoses to report and assigns billable ICD-10-CM codes as specific as the record allows, verified via the ICD-10 connector in the healthcare plugin. US outpatient; certified coder review required.

Anthropic's official skill for US prior auth review: checks records, provider credentials and codes, weighs medical necessity against Medicare policy, recommends approve, deny or pend and drafts the letter. Needs the healthcare plugin's CMS, NPI and ICD-10 connectors; clinicians confirm.

Anthropic's official healthcare plugin for the US system: extract data from clinical notes, assign ICD-10 and CPT/HCPCS codes, review prior auths, screen for fraud and draft trial protocols, with hosted CMS, NPI and PubMed connectors. FHIR access needs API rights; clinicians must review.

Deploys HTML or full-stack projects generated by agents to Tencent Cloud EdgeOne Makers and returns a public URL; supports both the international and China sites. Maintained by Tencent Cloud EdgeOne (formerly EdgeOne Pages MCP); sign in via browser or set an API token.

Creates a blueprint-style animation that explains UX decisions step by step for case studies and posts: rebuilds changed parts between Before and After screens, or annotates each module of one screen. Reproduces Figma frames 1:1; CC BY-NC 4.0, non-commercial only.

Official FinLab quant skill: lets agents use FinLab data (mainly Taiwan stocks, plus US, Japan, Korea and Hong Kong) to write stock-picking strategies, run backtests and analyze factors. Needs a FinLab account (free tier has a daily data cap), Python 3.10+ and uv. Not investment advice.

Makes a product film (landing-page loop, launch video, promo, demo reel) in code with Remotion, reusing the product's own design system, components and voice. It interviews the user first, then writes the story, syncs scenes to music beats, reviews stills in loops and verifies the final render.

Lets agents run local ComfyUI workflows to generate images, audio and video, regenerate from earlier results with new parameters, check the queue and switch models. Install and start ComfyUI, then clone this repo and run the server with Python (port 9000 by default).

Resend's official email CLI: send email and manage domains, API keys, contacts, broadcasts and webhooks, with automatic JSON output outside a terminal and a bundled agent skill. Needs a Resend API key; a free plan is available.

Spotify's official plugin: agents script and voice content such as morning briefings, meeting recaps or article explainers into audio episodes with cover art and a timeline, saved to Spotify. Needs the save-to-spotify CLI and sign-in; voices can use free local Kokoro.

One learn-faster command turns Claude Code into a learning coach: builds a syllabus for your level, schedules spaced review, sets exercises and tracks progress, with an exam-prep mode. Needs uv and Claude Code; Codex: run learn-faster init --agent codex (lighter onboarding).

Generates and edits images with Google's Gemini image model (Nano Banana) at up to 4K, keeping multiple characters consistent, rendering accurate text and setting aspect ratios. Needs a Gemini API key (or Vertex AI) and uv. Community project, not by Google.

Official Databricks skills guiding agents to build Lakeflow pipelines and jobs, Apps, Lakebase, Vector Search, Model Serving and AI/BI dashboards on the lakehouse. Needs the Databricks CLI.

Official Databricks plugin with skills for the CLI and auth, Genie data queries, Apps, Lakebase, Lakeflow jobs and pipelines, Model Serving and DABs deployment, plus hooks for prompt routing, session context and auth failure hints. Needs a Databricks account.

Translates SRT subtitles or plain text in batches through any OpenAI-compatible API, stripping timestamps and other overhead to save tokens, keeping output line-aligned with the source, and resuming interrupted jobs. Needs Node.js 20+, a source install and an OpenAI API key or local Ollama.