Spark
Sign In
MCP Connector

Automate GNURadio Flowgraph Creation

LLM-driven RF/SDR assistant for Claude Code that surveys spectrum, builds receivers, and saves reproducible GNU Radio projects.

Connect
Works with githubpython
Julius
Own this? Claim it
77
Spark score
out of 100
Updated last month
Version V1
Models
claude
Add to Favorites

Why it matters

Programmatically create, modify, and save GNURadio flowgraphs (.grc files) using an MCP server. This asset is designed for AI-driven automation and integration with LLMs.

Outcomes

What it gets done

01

Generate GNURadio flowgraphs from code or automation scripts.

02

Edit and save existing .grc files programmatically.

03

Integrate with AI and automation tools for dynamic flowgraph management.

04

Facilitate automated testing and debugging of GNURadio applications.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/vb-gnuradio | bash

Overview

Gnuradio MCP Server

Marconi is an LLM-driven RF assistant for Claude Code that surveys spectrum, builds and validates receivers via GNU Radio, and saves reproducible SigMF/YAML/.grc artifacts from plain-language descriptions. Use it for simulation-based RF exploration and receiver building; it's early-stage and simulation-only, with no real hardware SDR support in v1.0.

What it does

Marconi (formerly known as GNURadio/GR-MCP) is an LLM-driven RF assistant for Claude Code, built on GNU Radio. Describe what you want on the air in plain language, and it surveys the spectrum, builds and runs a receiver, inspects the signal, and leaves behind a reproducible RF project - SigMF captures, YAML pipelines, and .grc flowgraphs you own. It is currently early-stage and simulation-only in v1.0 (no hardware support yet).

It picks the right skill for the task and drives the workflow end to end, dropping generated artifacts under artifacts/ in the server's working directory (configurable via MARCONI_WORKSPACE).

When to use - and when NOT to

Use it when you want to explore RF signals conversationally - simulating a broadcast station and building a receiver for it, or surveying an existing IQ capture to identify and decode what's on the air - without hand-writing GNU Radio flowgraphs.

Do not use it for real hardware SDR workflows yet, since v1.0 is simulation-only. It also requires GNU Radio 3.10+ installed system-wide to run pipelines or simulations (analyzing existing IQ files alone doesn't need it).

Inputs and outputs

Inputs are natural-language descriptions of an RF scenario (e.g. a station frequency and modulation) or an existing IQ capture file to analyze. Outputs are SigMF captures, YAML pipeline definitions, and .grc GNU Radio Companion flowgraphs saved as reproducible artifacts, plus the receiver's decoded/demodulated output.

Capabilities

  • survey-spectrum: identify what's present via PSD, signal detection, and spectrogram analysis
  • simulate-scene: register a simulated device from a description
  • build-receiver: channelize, demodulate, validate, run, verify, and save a receiver
  • debug-no-signal: locate the fault when a receiver outputs nothing
  • tx-experiment: closed-loop transmit-then-receive in simulation
  • escape-hatch: drop to the underlying Python library when no built-in tool fits

How to install

Requires uv and Python 3.13+, plus GNU Radio 3.10+ installed system-wide (via conda-forge, your distro's package manager, or Homebrew) for running pipelines and simulations. In Claude Code:

/plugin marketplace add yoelbassin/gr-mcp
/plugin install marconi

uv starts the MCP server on first use and loads the skills automatically - nothing else to configure.

Who it's for

RF/SDR hobbyists and engineers who want to prototype and analyze radio signals conversationally with GNU Radio, without manually assembling flowgraphs.

Source README

Marconi 🤖

        o)))
        |
    .-------.
    | o   o |
    |   v   |
    '-------'

    Hi! I'm Marco.

LLM-driven RF for Claude Code. Describe what you want on the air and I survey the
spectrum, build and run the receiver, look at the signal, and leave a reproducible
RF project behind - SigMF captures, YAML pipelines, and .grc flowgraphs you own.

Early development - v1.0 is simulation-only (no hardware yet). See ROADMAP.md.

Proviously known as GNURadio/GR-MCP.

Requirements

  • uv and Python ≥ 3.13
  • GNU Radio 3.10+ installed system-wide (conda-forge,
    your distro's package manager, or Homebrew) - needed to run pipelines and
    simulate. Analyzing existing IQ files doesn't require it.

Install

In Claude Code:

/plugin marketplace add yoelbassin/gr-mcp
/plugin install marconi

That's it - uv starts the MCP server on first use and loads the skills. Nothing
else to configure.

Use it

Ask in plain language:

"Simulate an FM station at 100.3 MHz and build me a receiver."

"Here's a capture, mystery.wav - survey what's on the air, then decode the strongest carrier."

I pick the right skill and drive the workflow end to end:

  • survey-spectrum - what's present (PSD, signal detection, spectrogram)
  • simulate-scene - register a simulated device from a description
  • build-receiver - channelize, demodulate, validate, run, verify, save
  • debug-no-signal - locate the fault when a receiver outputs nothing
  • tx-experiment - closed-loop transmit-then-receive in simulation
  • escape-hatch - drop to the Python library when no tool fits

Artifacts land under artifacts/ in the server's working directory; set
MARCONI_WORKSPACE to put them elsewhere.

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.