radio-copilot

What This Skill Does

The radio-copilot skill is a specialized, zero-AI orchestration tool designed to bridge the gap between satellite orbital mechanics and amateur radio software-defined radio (SDR) operations. Instead of relying on non-deterministic neural networks, it uses precise TLE (Two-Line Element) calculation to predict satellite passes for NOAA APT, METEOR LRPT, and the ISS. It handles the scheduling math for you, determining exactly when a satellite will rise (AOS) and set (LOS) based on your specific geocoordinates. The skill provides actionable telemetry, including azimuth and elevation start/end points, track direction, and orbital inclination, formatted specifically for manual dish alignment or automated tracking systems. Furthermore, it includes optional hooks for remote hardware integration, allowing you to trigger a Raspberry Pi for signal capture and a Jetson device for downstream decoding using SatDump, creating a fully automated satellite reception pipeline.

Installation

To install the radio-copilot, use the ClawKit CLI within your OpenClaw environment. Run: clawhub install openclaw/skills/skills/davestarling/satellite-copilot. Post-installation, you must initialize your configuration directory by running mkdir -p ~/.clawdbot/radio-copilot. Copy the provided config.example.json into this folder, ensuring strict file permissions with chmod 600. The orchestrator requires a periodic trigger, which is best achieved by adding the orchestrator.py script to your system crontab (e.g., */5 * * * *). This ensures your WhatsApp alerts for upcoming passes are always current and based on the most recent NORAD TLE data.

Use Cases

1. Manual Satellite Tracking: Hobbyists using handheld Yagi antennas or manual rotators receive timely alerts before a pass starts, including precise azimuth and elevation coordinates to manually track the satellite across the sky.
2. Automated SDR Capture: Users running dedicated SDR stations can configure the skill to trigger remote capture scripts on a Raspberry Pi, enabling 'set it and forget it' imagery collection from NOAA weather satellites.
3. Post-Processing Automation: Integrate with a Jetson Nano or similar compute device to automatically run SatDump pipelines as soon as the capture file is finalized, turning raw I/Q data into processed imagery without human intervention.

Example Prompts

1. "Check for the next NOAA-19 pass at my current coordinates and send a WhatsApp alert with the AOS elevation."
2. "Update my radio-copilot configuration to only track satellites with a minimum elevation of 20 degrees."
3. "List all upcoming ISS passes for the next 24 hours and include the track direction for each."

Tips & Limitations

• Zero-AI Reliability: Because this tool uses deterministic orbital mechanics, the alerts are significantly more reliable than AI-generated schedules. Trust the script's output over generative estimations.
• TLE Data: Ensure your system has internet access periodically; if the TLE data becomes stale, your pass calculations will drift significantly and result in missed captures.
• Hardware Hooks: Keep your capture and decode scripts modular. If the SatDump decode fails, ensure your logic provides a fallback to save the raw capture file for manual debugging later.
• Geographic Precision: Always verify your latitude and longitude in the config file. Even a few miles of error can drastically shift the perceived azimuth and elevation for low-orbit satellites.