skill

sdr-skills

In-depth Software Defined Radio (SDR) agent skill: DSP, modulation, hardware, GNU Radio, protocol decoding, signal identification, IQ/SigMF, antennas, regulations, and 8 runnable Python tools.

briannasywa/sdr-skillscommunitylistedunreviewed

Use skill

Set up

Skills are installed by placing their folder where your agent loads skills.

Download ZIP from GitHub Open the repository

The download comes from github.com/briannasywa/sdr-skills itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 23, 2026

SDR Skills

An in-depth Agent Skill for Software Defined Radio (SDR). It gives an AI coding agent (Claude Code, and other Agent-Skills-compatible tools) deep, practical knowledge of receiving, analyzing, decoding, and lawfully transmitting radio signals with SDR hardware such as RTL-SDR, HackRF, Airspy, SDRplay, ADALM-Pluto, LimeSDR, BladeRF, and USRP.

What is in it

One skill, software-defined-radio, built for progressive disclosure: a compact SKILL.md router plus fourteen deep reference files, eight runnable Python tools, and GNU Radio templates.

Area	File
DSP fundamentals (sampling, IQ, FFT, filters, decimation, sync, SNR)	`references/dsp-fundamentals.md`
Modulation and demodulation (AM/FM/SSB/FSK/PSK/QAM/OFDM/LoRa/DSSS)	`references/modulation.md`
Hardware (every common SDR, specs, quirks, front-end accessories)	`references/hardware.md`
Software toolchain (drivers, receivers, decoders, install recipes)	`references/software-toolchain.md`
GNU Radio 3.10 (blocks, flowgraphs, parameters, debugging)	`references/gnuradio.md`
Python DSP (numpy/scipy, SoapySDR, pyrtlsdr, pyadi-iio, UHD, SigMF)	`references/python-sdr.md`
Protocol decoding catalog (ADS-B, AIS, ACARS, POCSAG, DMR, satellites, LoRa, rtl_433, GSM...)	`references/protocols.md`
Signal identification (procedure + shape/band lookup tables)	`references/signal-identification.md`
Antennas and RF front-end (link budget, LNA, coax, noise)	`references/antennas-rf-frontend.md`
IQ formats and SigMF (cu8/cs8/cs16/cf32, metadata)	`references/iq-formats.md`
Spectrum regulations (ITU, FCC, Ofcom, ETSI, Indonesia Komdigi)	`references/regulations.md`
Troubleshooting (device, overruns, ppm, DC, overload)	`references/troubleshooting.md`
Transmission and lab safety (authorized TX, cabled replay)	`references/transmission-lab-safety.md`
Advanced topics (DF, passive radar, GNSS, RFML, RFNoC)	`references/advanced-topics.md`

Bundled tools (`scripts/`, Python 3.9+, numpy)

Script	Purpose
`sdr_calc.py`	Wavelength, dipole, dBm/W/uV, noise floor, FSPL, link budget, Friis NF, Carson BW, FFT resolution, Doppler, file size, symbol-rate/BW
`iq_info.py`	Detect format, DC offset, clipping, level, occupied bandwidth, peak frequency
`iq_spectrum.py`	PSD + waterfall PNG (ASCII fallback when matplotlib is absent)
`iq_convert.py`	Convert cu8/cs8/cs16/cf32/cf64, with optional shift/decimate/trim
`sigmf_write.py`	Write a `.sigmf-meta` sidecar for an existing capture (no dependencies)
`fm_demod.py`	WBFM/NBFM/AM demodulation from IQ to WAV
`symbol_rate.py`	Blind symbol-rate and modulation hints for digital bursts
`probe_devices.py`	Enumerate attached SDRs via rtl_test/hackrf_info/SoapySDRUtil/uhd/iio

scipy and matplotlib are optional; every script degrades gracefully with numpy alone.

Install

With the skills CLI:

npx skills add briannasywa/sdr-skills --skill software-defined-radio

Or copy the skill folder into your agent's skills directory:

# Claude Code (project-local)
cp -r skills/software-defined-radio .claude/skills/
# or user-global
cp -r skills/software-defined-radio ~/.claude/skills/

The skill activates automatically when you mention SDR, RTL-SDR, HackRF, GNU Radio, IQ samples, a waterfall, demodulation, a radio protocol, or an RF capture.

Use the tools directly

cd skills/software-defined-radio
python scripts/sdr_calc.py dipole --f-mhz 137.5
python scripts/sdr_calc.py noise --bw 2e6 --nf 6 --snr 12
python scripts/iq_info.py capture.cu8 --fs 2.4e6 --fc 433.92e6
python scripts/iq_spectrum.py capture.cu8 --fs 2.4e6 --fc 100e6 -o spec.png
python scripts/probe_devices.py

Scope and safety

This skill is receive-first. It helps with listening, signal identification, decoding open broadcasts (aviation, maritime, weather satellites, amateur, unencrypted telemetry), DSP, and lawful transmission (licensed amateur, unlicensed bands with certified devices, or cabled lab setups on your own equipment). It will not help defeat access control, spoof or jam, intercept private or cellular communications, or transmit where doing so is illegal. See references/regulations.md and references/transmission-lab-safety.md.

Requirements

Skill knowledge: none; it is Markdown plus Python.
Tools: Python 3.9+ with numpy. scipy and matplotlib recommended. SDR command-line tools (rtl_sdr, hackrf, SoapySDR, GNU Radio) only for live hardware work; install recipes are in references/software-toolchain.md.

Contributing

Issues and pull requests welcome: corrections, new protocol entries, additional hardware, regulation updates for more countries. Keep reference files focused and under ~250 lines, keep SKILL.md under 500 lines, and validate with skills-ref validate skills/software-defined-radio.

License

MIT. See LICENSE.