---
name: al-645-run-a-complex-instrument-from-one-plain-english-ask-let-a
description: "NVIDIA’s Oct 30, 2026 blog covers Breakthrough Listen and the SETI Institute at the Allen Telescope Array: agentic AI on the Holoscan platform lets researchers steer antennas, calibrate, correct live data and rewrite signal-processing blocks from a single natural-language request. Nexus is the observatory orchestrator (agent harness); DAQIRI streams sensor data into GPU memory; CyberEther does CUDA-X signal processing. NVIDIA cites up to 2.4 Tbps on RTX PRO 6000 Blackwell GPUs on premises, and the pipeline’s first real-time AI detection of a Crab Nebula pulsar. Andrew Siemion (SETI Institute / Breakthrough Listen) frames agents as speeding a search that may increasingly listen for other agents. Use when the user wants help with natural-language telescope ops agent (research) for a small business."
---

# Run a complex instrument from one plain-English ask: let an agent steer the hardware, clean the live data stream, fix errors and plot the result while you watch

NVIDIA’s Oct 30, 2026 blog covers Breakthrough Listen and the SETI Institute at the Allen Telescope Array: agentic AI on the Holoscan platform lets researchers steer antennas, calibrate, correct live data and rewrite signal-processing blocks from a single natural-language request. Nexus is the observatory orchestrator (agent harness); DAQIRI streams sensor data into GPU memory; CyberEther does CUDA-X signal processing. NVIDIA cites up to 2.4 Tbps on RTX PRO 6000 Blackwell GPUs on premises, and the pipeline’s first real-time AI detection of a Crab Nebula pulsar. Andrew Siemion (SETI Institute / Breakthrough Listen) frames agents as speeding a search that may increasingly listen for other agents.

## Inputs to collect first
Ask the user for anything below that they haven't already given you. Don't guess missing facts.

- Target

## Instructions

Do the task below with the user's inputs in place of each [bracket].

1. Name the instrument or lab rig and the three actions you repeat most (point/aim, calibrate, plot or export).
2. Give an agent read access to live status and a short allow-list of safe commands; deny anything that can damage hardware without a human confirm.
3. Prompt pattern: ‘Aim at [target], calibrate for a clean signal, flag packet loss, and plot the spectrogram.’
4. Keep processing next to the instrument when data volume or locality matters; log every command the agent ran.
5. Start in suggest mode (agent proposes the command sequence, you approve), then allow auto-run only for the safe subset.
6. Measure time from request to usable plot versus the old manual checklist.

## Safety checks (share these with the user)

- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
- Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed.

## About this skill

- Works with: NVIDIA Holoscan, Nexus observatory orchestrator, DAQIRI, CyberEther, CUDA-X, on-prem GPUs
- Based on this real-world example: https://blogs.nvidia.com/blog/ai-joins-the-search-for-extraterrestrial-signals/
- Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-645-run-a-complex-instrument-from-one-plain-english-ask-let-an-a/
- To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/