Herramientas

High-throughput LLM inference and serving engine: vllm serve starts an OpenAI- and Anthropic-compatible server in one command. Supports 200+ model architectures from Hugging Face, quantization and distributed inference. Aimed mainly at NVIDIA, AMD and other GPUs.

Provides a sequential_thinking tool that guides the model to break complex problems into revisable, branching steps. Good for planning, analysis and tasks that need careful iteration. Official MCP reference server.

Baidu PaddlePaddle's official OCR and document parsing tool: the paddleocr command reads text from images and parses PDFs and scans into Markdown or JSON, supporting 100+ languages with models that run locally.

Official PaddleOCR MCP: text recognition and layout parsing for images and PDFs (PP-OCR, PP-StructureV3, PaddleOCR-VL), returning text or Markdown. Runs inference locally, or calls the official API or a self-hosted service.

Astral's extremely fast Python package and project manager written in Rust, replacing pip, pipx, poetry, pyenv and more. Supports lockfiles and multiple Python versions, and uvx runs Python tools without installing them.

MCP server built into the adaptive Python scraping framework Scrapling: supports plain HTTP requests, browser rendering and stealth fetching, and uses CSS selectors to pass only the target content to the AI, saving tokens and filtering out prompt injection.

Adaptive web scraping framework: scrapling extract turns a page into Markdown, text or HTML in one command, with CSS selectors and browser rendering for dynamic pages, plus an MCP server (needs the ai extra). Needs Python 3.10+ and a browser install on first use.

Open-source, LLM-friendly web crawler that turns pages into clean Markdown for RAG and agents. Ships the crwl CLI and supports deep crawling and question-driven data extraction (extraction needs an LLM key). Free to self-host.

Document parser from OpenDataLab that turns PDFs, images and Office files into Markdown or JSON, handling tables, formulas and complex layouts well, with page-by-page reading and citation anchors for agents. Parses locally by default; needs Python 3.10+.

Classic open-source OCR engine and CLI that reads text from images and outputs plain text, hOCR, searchable PDF or TSV. Supports 100+ languages; install the language data for each language you need.

All-in-one LLM fine-tuning framework: llamafactory-cli handles pretraining, SFT, PPO/DPO alignment and LoRA/QLoRA training, plus chat inference, weight merging and export, and OpenAI-style API serving for 100+ open models. Has a Web UI; training usually needs a GPU.

Multi-agent orchestration framework for Claude Code and Codex, formerly Claude Flow: swarm coordination, self-learning memory and cross-machine federated communication. The marketplace offers 30+ plugins; full features need npx ruflo init.

Multi-agent orchestration framework formerly called Claude Flow: ruflo init gives Claude Code nearly 100 specialized subagents, swarm coordination, cross-session vector memory and hooks; add --codex to configure Codex instead. Also works as a standalone MCP server or plugin marketplace.

Job search system that runs in your local AI coding CLI: scans job portals, scores roles with a structured A-H evaluation and a 1-5 rating, flags fake listings, tailors ATS resumes and cover letters, and tracks applications. Never submits for you. Needs Node.js.

Parses PDF, DOCX, PPTX, XLSX, HTML, images and audio, recovering layout, reading order, tables and formulas, and exports Markdown, HTML or lossless JSON. Started by IBM Research, now hosted by the LF AI & Data Foundation; runs fully locally.

Open-source agentic video production: open Claude Code or Codex in the cloned repo, describe the video, and the agent handles research, script, voiceover, footage, subtitles and editing, confirming storyboard and cost first. Needs Python, Node.js and FFmpeg; free assets work without API keys.

Command-line toolkit for transcoding, cutting, joining, frame extraction and streaming of audio and video, including ffmpeg, ffprobe and ffplay. A core dependency for agents handling media. The GitHub repo is the official mirror.

HeyGen's official HyperFrames skill set: lets agents write deterministic MP4 videos in HTML, CSS and animation, with workflows for promo videos, explainers, captions and beat-synced music. Needs Node.js 22+ and FFmpeg.

Write videos in HTML, CSS and frame-seekable animations, preview them from the CLI and render deterministic MP4s. Comes with agent skills for product launch videos, captions and talking-head re-edits. Maintained by HeyGen; needs Node.js 22+ and FFmpeg.

HeyGen's official video plugin: compose scenes with HTML, CSS and animation and render deterministic MP4s, with skills for promo videos, captions, voiceover and music, and PR walkthrough videos. Needs Node.js 22+ and FFmpeg.

Write videos in React and render them from the command line with npx remotion render, letting agents batch-produce data-driven videos. Free for individuals, nonprofits and companies of up to 3 people; larger for-profit companies need a license.

Open-source crypto trading framework: download market data, backtest and optimize strategies, and paper trade from the CLI while agents iterate on strategies. Docker recommended; strategies need Python and live trading needs exchange API keys. Backtests don't guarantee returns.

General-purpose AI agent that runs on your machine, with a desktop app, CLI and API. Written in Rust, works with 15+ model providers and connects to 70+ extensions via MCP. Originally from Block, now part of the Linux Foundation's AAIF.

C/C++ port of OpenAI's Whisper speech recognition model that transcribes audio offline and outputs subtitles. Runs on CPU, with Apple Silicon and various GPU acceleration. Download a ggml-format model first.