MCP.so
Sign In

open-web-agent-rs

@seemueller-io

Repository

About open-web-agent-rs

an mcp server with candle inference

#agent#rust#axum#llm#genaiscript

Basic information

Category

AI & Agents

License

MIT license

Runtime

rust

Transports

stdio

Publisher

seemueller-io

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "open-web-agent-rs": {
      "command": "docker",
      "args": [
        "compose",
        "up",
        "-d",
        "searxng"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is open-web-agent-rs?

open-web-agent-rs is a Rust-based web agent that includes an embedded OpenAI‑compatible inference server. It supports only Gemma models and is designed for developers who want a lightweight, local web agent with custom inference.

How to use open-web-agent-rs?

Copy the .env.example to .env, install dependencies with bun i, start the local inference server by running cargo run --release -- --server inside local_inference_server, launch the SearXNG search engine with docker compose up -d searxng, then run bun dev to start the agent.

Key features of open-web-agent-rs

  • Rust‑based implementation for performance and reliability
  • Embedded OpenAI‑compatible inference server
  • Supports Gemma models only
  • Integrates with SearXNG for web search
  • Quick setup with Bun, Cargo, and Docker

Use cases of open-web-agent-rs

  • Creating a private web agent with local Gemma model inference
  • Experimenting with Rust‑based agent workflows
  • Building a self‑hosted, search‑enabled assistant

FAQ from open-web-agent-rs

Which models does open-web-agent-rs support?

It supports only Gemma models. Other model families are not compatible.

What are the runtime requirements?

You need Rust, Bun, and Docker installed. The local inference server is written in Rust and runs via Cargo.

Where does data live?

Data can be configured in the .env file. The SearXNG instance runs locally via Docker.

Are there any known limitations?

The inference server is embedded and supports only Gemma models, so flexibility is limited to that model family.

What transport or authentication does it use?

The README does not specify authentication or transport details. It uses an OpenAI‑compatible API interface locally.

Comments

More AI & Agents MCP servers

🔎 GPT Researcher

assafelovic

An autonomous agent that conducts deep research on any data using any LLM providers

28.1K

Web Agent Protocol

OTA-Tech-AI

🌐Web Agent Protocol (WAP) - Record and replay user interactions in the browser with MCP support

500

Sequential Thinking Multi-Agent System (MAS)

FradSer

An advanced sequential thinking process using a Multi-Agent System (MAS) built with the Agno framework and served via MCP.

306

meGPT - upload an author's content into an LLM

adrianco

Code to process many kinds of content by an author into an MCP server

287

Unreal Engine Generative AI Support Plugin

prajwalshettydev

Unreal Engine plugin for LLM/GenAI models & MCP UE5 server. OpenAI GPT-5, Deepseek R1, Claude Opus/Sonnet, Gemini 3, Grok 4, Alibaba Qwen, Kimi, ElevenLabs TTS, Inworld, OpenRouter, Groq, GLM, Ollama, Local, Meshy, Tripo, Hunyuan3D, Rodin, fal, Dashscope, Seedream. NPC AI, agenti

622