// the find
fetchai/innovation-lab-examples
80+ production-ready AI agent examples in Python — build autonomous agents, multi-agent systems and agentic AI with uAgents, ASI:One, MCP, A2A, LangChain, CrewAI, Gemini, Claude and OpenAI.
A mono-repo of 80+ independent example projects showing how to build agents on Fetch.ai's stack (uAgents, Agentverse, ASI:One), plus integrations with LangChain, CrewAI, A2A, MCP, and the usual LLM vendors. It's aimed at hackathon participants and developers evaluating Fetch.ai's agent tooling who want a working starting point rather than docs alone.
Every example is genuinely self-contained — own README, requirements.txt, .env.example — so you can clone one folder's worth of code without untangling shared state from the rest of the repo. There's a real CI setup (per-example test runner, ruff lint baseline, PR review workflow) which most sprawling examples repos skip entirely. Coverage is wide: payment flows (Stripe, Skyfire, FET), A2A agent-to-agent protocol, MCP servers, and multi-framework orchestration (CrewAI, AG2, Google ADK) all have working code, not just descriptions.
The 'production-ready' claim in the description doesn't match the content — these are demo-depth scripts wired to one vendor SDK each, not things you'd run against real traffic or real money without rewriting the error handling and retry logic. Most examples pull in uAgents/Agentverse even when the actual integration being demonstrated (e.g. a LangChain or CrewAI pattern) has nothing to do with Fetch.ai's framework, which makes it harder to lift the useful part into a project that isn't already on their stack. With 80+ folders and an open contributors/ directory, there's no visible signal for which examples are maintained versus abandoned after the hackathon that spawned them — you're on your own to figure out which ones still work against current API versions.