tinyfish-io/tinyfish-cookbook

A collection of sample apps and recipes built with the TinyFish web agent. Open-source examples for you to learn & build!

What it solves

TinyFish acts as a bridge between the live web and AI agents. It solves the problem of accessing websites that lack official APIs, allowing developers to extract structured data or perform complex browser-based tasks without manually writing scrapers or managing browser infrastructure.

How it works

TinyFish provides a set of API endpoints that handle web interactions for you. You can perform simple searches, fetch clean content from specific URLs in formats like Markdown or JSON, or provide a natural-language goal to have an agent navigate a site and return structured results. It includes built-in features like stealth browsing, session management, and token-efficient content filtering.

Who it’s for

This repository is for developers building AI agents, automation scripts, or research tools who need reliable, programmatic access to web content. It is designed for anyone who wants to turn websites into programmable data sources.

Highlights

  • Offers free Search and Fetch endpoints with generous rate limits.
  • Supports natural-language goals for multi-step browser automation.
  • Provides built-in stealth profiles and rotating proxies to handle dynamic content.
  • Includes a "Vault" for managing agent-grade credentials and session memory.
  • Features a collection of "recipes" and demos for use cases like price comparison, market intelligence, and research automation.
  • Compatible with various tools via REST API, CLI, SDKs (Python/TypeScript), and MCP servers.

Related

  • Project
    aattaran/deepclaude

    A proxy tool that lets you use Claude Code's autonomous agent loop with cheaper model backends like DeepSeek V4 Pro, OpenRouter, or Fireworks AI.

  • Project
    kunchenguid/axi

    AXI (Agent eXperience Interface) is a framework and catalog of token‑efficient CLI wrappers built for AI agents. It defines ten design principles that keep output small, deterministic, and easy for LLMs to parse, and provides ready‑made tools (e.g., `gh-axi`, `chrome-devtools-axi`) that outperform standard CLIs in cost, speed, and success rate.

  • Project
    mgechev/skills-best-practices

    A comprehensive guide on best practices for creating structured agent skills to improve AI agent reliability, discoverability, and token efficiency.

  • Project
    InternLM/lagent

    A PyTorch-inspired framework for building multi-agent AI applications with support for synchronous and asynchronous workflows and integrated tool execution.