도구

Skyvern's official MCP: drives a browser with vision LLMs to complete multi-step tasks like filling forms, downloading files and web research. Connect to Skyvern Cloud (OAuth or API key), or self-host Skyvern and run it as a local process.

Lets agents look up docs, properties and workflow templates for n8n's 2,000+ nodes and build and validate workflows; with an n8n API key configured it can create and update workflows in your instance directly.

ML-based video upscaling and frame interpolation using Real-ESRGAN, Real-CUGAN, RIFE and other models to upscale old low-res video or raise frame rates; video2x handles one video per run and is scriptable. Needs a Vulkan GPU; Windows, Linux and Docker builds, no native macOS.

Offline speech transcription and translation based on Whisper; audio never leaves your machine. buzz add exports lectures, interviews and meeting recordings to TXT, SRT or VTT. pip install needs Python 3.12 and FFmpeg; a desktop app is also available. Check transcripts manually.

Continuously records your screen and audio locally, extracting on-screen text and transcribing speech, so agents can search pages, docs and meetings you saw and summarize your day. Data stays local. Needs screenpipe running: source is free for personal use, desktop app is a subscription.

Compiles Markdown chapters into an online book with a table of contents, full-text search and themes, good for manuals, lecture notes and course material; agents can edit chapter files and rebuild to preview. Maintained by the Rust team; install a prebuilt binary or via cargo.

Scripts terminal sessions in .tape files and records them to GIF, MP4 or WebM, great for CLI demos and docs; text snapshot output also enables terminal integration tests. By Charm; requires ttyd and ffmpeg.

Turns a project directory or website URL into a roughly 20-second launch video: gathers copy, colors and real UI from the code or live site, storyboards it, and builds visuals, music and the render with tools already on the machine, delivering an MP4, poster frame and share copy.

Industrial-grade open-source speech recognition toolkit from Alibaba's ModelScope, with an agent-friendly funasr command that turns audio into text, JSON or SRT subtitles in one step, with speaker diarization and timestamps. Strong on Chinese. Requires PyTorch.

Official MCP from the FunASR project: transcribes audio files locally with the SenseVoice model, supporting Mandarin, Cantonese, English, Japanese and Korean, with optional segment timestamps. Local inference needs no API key and audio stays on your machine. Docker image available.

Turns the current project into a 15-25 second shareable launch video: reads the code, plans a concept and storyboard, hands a composition brief to Hyperframes to render with music and sound effects, and writes share copy. Supports tone, format and duration options plus optional Kokoro voiceover.

Command-line tool for bulk-downloading galleries and posts from many image hosts and social platforms, with flexible file naming and config, and yt-dlp integration for video. Main development has moved to Codeberg; the GitHub repo stays in sync.

Unofficial Python library and CLI for driving Google NotebookLM (now Gemini Notebook): bulk-import sources, ask cited questions, and generate and download podcasts, videos, slides, quizzes and flashcards. Ships an agent skill. Relies on undocumented APIs and needs a Google sign-in.

Pressures the agent with big-tech PUA and PIP-style rhetoric plus structured debugging steps and initiative checks, pushing it to change approach after failures and show evidence before claiming done. Modes include always-on, encouragement and P9 lead. Works with Claude Code, Codex and more.

AI2's tool for converting PDFs and scans to Markdown with a vision LLM, handling equations, tables, handwriting and multi-column layouts, suited to bulk digitizing papers and archives. Local runs need an NVIDIA GPU with 12GB+ VRAM; otherwise use a paid OpenAI-compatible remote endpoint.

All-in-one video translation and dubbing: chains speech recognition, subtitle translation, multi-speaker AI voiceover and audio-video muxing, with local or online models like Faster-Whisper, DeepSeek and Edge-TTS. cli.py supports headless batch runs. Needs FFmpeg.

Reports token usage and cost for Claude Code, Codex, Gemini CLI and other coding agents from local logs, by day, week, month or session, and shows Claude Code's 5-hour billing windows.

Define and run chains of HTTP requests in plain text files, capture variables and assert on status codes, headers and JSONPath or XPath results, for both fetching data and API testing in CI. Open source from Orange.

Browser control from the browser-use team: an editable CDP connection lets agents drive the Chrome you are already using, writing missing helper functions as they go and keeping them for reuse. Also offers an MCP server. Allow remote debugging in Chrome on first use.

Lets Claude Code push ML research forward unattended: literature review, topic selection, experiment deployment, cross-model review loops and paper writing through rebuttal, in 80+ skills. Codex acts as the independent reviewer by default; other model pairings work too.

Strips AI-sounding prose from English drafts: lists phrases to cut and patterns to avoid, then scores directness, rhythm, trust in the reader, authenticity and density, sending anything under 35/50 back for revision. Word lists and examples are English, for blogs, emails and articles.

Turns Claude Code or Codex into a motion studio: from your site or product it writes storyboards, animates shots and adds sound, rendering promo videos with Remotion. Edit further in a browser workspace or export JianYing drafts. Needs Node.js and ffmpeg; Remotion may need a paid license.

Sealos's Kubernetes lifecycle CLI: one sealos run deploys a highly available cluster across multiple servers, and cluster images install databases, middleware and other apps, with offline delivery support. Linux only.

Context and harness engineering skills covering context compression and optimization, multi-agent architectures, memory systems, tool design and LLM-as-judge evaluation, activated automatically by task. Aimed at building production-grade agent systems.