> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/relay/_mcp/server.

# gRPC Worker Plugins

## Docs

- [About gRPC Worker Plugins](https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/about.md): Choose a Python, Rust, or custom grpc-v1 worker implementation.
- [Python Worker](https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/python.md): Build, package, activate, verify, and remove the checked Python worker.
- [Rust Worker](https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/rust.md): Build, digest, activate, verify, and stop the checked Rust grpc-v1 worker.
- [Middleware and Continuations](https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/middleware-and-continuations.md): Implement every worker middleware closure and downstream continuation path.
- [Runtime Events and Scopes](https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/runtime-events-and-scopes.md): Use worker host-runtime marks, scopes, stack binding, and cleanup.
- [grpc-v1 Protocol Reference](https://docs.nvidia.com/nemo/relay/v0.9.4/build-plugins/workers/grpc-v1-protocol.md): Reference every worker RPC, registration surface, envelope, capability, and shutdown stage.