
pydantic-ai
AgentType-safe Python framework for building agents and LLM applications
- Open source
- Agent framework
- 21.2K GitHub stars
- Design and media

Type-safe Python framework for building agents and LLM applications
MCP server that builds a complete one-page small business website from a short brief
Swap yourself for a fictional AI Stand-in in your phone videos, from upload to download

Pulls transcripts, chapters and metadata from videos so agents can read and cite them
Runs locallyPull every video of a YouTube channel with stats via the official API, pay per video
Search YouTube and pull trending charts with stats through your own API key
Bulk YouTube video stats and growth tracking from links or playlists via your own key

Generate, extend and fix video sound effects, and turn audio into MIDI, from chat
Not checked yetBook covers from a title and author: 39 styles, SVG, PNG and print-ready PDF export

Search and read Flood's React components, recipes, tokens and design guidance
67 pay-per-call AI tools for agents, settled in USDC or USDG over x402
MCP server to search stock footage shot by real people and license it
Not checked yetVoice AI from Cartesia: text-to-speech, transcription, voice cloning and pronunciation
Not checked yetMake vertical AI meme videos from one or two photos, from your assistant or API
Generate and edit images with 18 models, including GPT Image 2.6, Nano Banana and FLUX 3

Bulk OCR of scanned PDFs and images to text, one row per page, via Apify

Remote MCP server for researching and making ads and videos from an AI chat
Not checked yetAsk Claude, ChatGPT or Codex for a narrated demo video of your web app
Not checked yetGenerate images and video with several models through one KeelRouter API key
Reverse image search through Yandex: matching pages, similar images and shop products
Remote MCP server that turns STEP, DXF and SVG files into CNC and laser G-code
MCP endpoint for pay-as-you-go speech-to-text, text-to-speech, OCR and decision APIs
Remote MCP server to search a curated design archive and save finds to collections
Not checked yetSearch model prices, chat and generate images, video and audio via TokenLab
Runs locallySplits a flat image into up to 100 named transparent PNG layers, paid in credits
Changes or removes text in a screenshot or photo, keeping the original font and layout
Image and video generation across several models through a single MCP
Search 151 UI styles and lint AI-written code against a style's rules

Run AI video interviews and read evidence-linked summaries from your assistant
Generate images from your MCP client with TopxAI's image model, billed per call