agentleFS
Sign inStart free
Seeed-ProjectsSeeed-Projects/Seeed-Jetson-DevelopTool/skills/claude/lumeo-setup/CLAUDE.md

Seeed-Jetson-DevelopTool / lumeo-setup

CLAUDE.md57 starsChanged 4 months agoMITView on GitHub

Install and configure Lumeo no-code video analytics gateway on NVIDIA Jetson. Covers account creation, gateway deployment, camera linking, and building a people detection pipeline. Requires Jetson with JetPack 5.1+ and internet access.

  • Installs packages

The file

We split this CLAUDE.md into the rules it gives an agent

9 rules. About 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run one phase at a time. After each phase:

    line 12
  2. Very specific:

    If output contains [STOP] → stop immediately, consult the failure decision tree below.

    line 14
  3. Very specific:

    If output ends with [OK] → tell the user "Phase N complete" and proceed to the next phase.

    line 15
  4. Distinctive:

    Verify JetPack version and network connectivity.

    line 33
  5. Distinctive:

    Delete the default cloud gateway under Gateways (free accounts allow only one gateway).

    line 49
  6. Distinctive:

    Run the installer script on the Jetson device:

    line 57
  7. Distinctive:

    Verify the container is running:

    line 71
  8. Distinctive:

    Keep default blocks (people detection model, tracker, WebRTC encoder) and click "Deploy".

    line 104
  9. Distinctive:

    Wait for a green "running" status indicator.

    line 106

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Getting Started with Lumeo on NVIDIA Jetson

Execution model

Run one phase at a time. After each phase:

  • Relay all command output to the user.
  • If output contains [STOP] → stop immediately, consult the failure decision tree below.
  • If output ends with [OK] → tell the user "Phase N complete" and proceed to the next phase.

Prerequisites

RequirementMinimum
HardwareNVIDIA Jetson device (tested on reComputer J4012)
JetPack5.1+ with all SDK components installed
InternetRequired on both Jetson and host PC
Host PCWindows, Linux, or Mac with browser access
Lumeo accountFree account at https://console.lumeo.com/register

Phase 1 — Preflight

Verify JetPack version and network connectivity.

cat /etc/nv_tegra_release
ping -c 3 lumeo.com

Expected: L4T version shown, ping succeeds. [OK] when both pass. [STOP] if no network.

Phase 2 — Create Lumeo account (manual)

  1. Visit https://console.lumeo.com/register — sign up with email and password.
  2. Sign in at https://console.lumeo.com/login.
  3. Enter an organization name and workspace name, click "Start using Lumeo".
  4. Delete the default cloud gateway under Gateways (free accounts allow only one gateway).

[OK] when Lumeo console is accessible and default cloud gateway is deleted.

Phase 3 — Install Lumeo Gateway on Jetson

Run the installer script on the Jetson device:

bash <(wget -qO- https://link.lumeo.com/setup)

Respond to prompts (defaults are fine). When the installer completes successfully, type install at the prompt to deploy a new gateway container:

Enter the command: install

Enter a container name and log in with your Lumeo credentials when prompted.

Verify the container is running:

Enter the command: list

Type exit to quit the script.

# Verify gateway appears in Lumeo console under Gateways

[OK] when the gateway shows as online in the Lumeo console. [STOP] if installer fails or gateway does not appear.

Phase 4 — Add cameras to the gateway

  1. Connect a USB camera to the Jetson or ensure an ONVIF camera is on the same network.
  2. In Lumeo console, click "Add Camera" under your gateway.
  3. Click "Link" next to a discovered camera (or "Manually add cameras" if auto-discovery fails).
  4. Enter camera name, provide credentials if needed, click "Connect camera".
  5. Click the linked camera to verify a preview snapshot appears.

For RTSP/HTTP streams: navigate to Deploy > Streams, click "Add input stream".

[OK] when camera preview is visible in the Lumeo console.

Phase 5 — Build a people detection pipeline

  1. In Lumeo console, go to Design > Solutions, select "Basics - Detect Objects".
  2. Keep default blocks (people detection model, tracker, WebRTC encoder) and click "Deploy".
  3. Select your Jetson gateway, choose the configured camera, click "Deploy".
  4. Wait for a green "running" status indicator.
  5. Click the play button to view the output stream via WebRTC.

[OK] when the live detection stream is visible with bounding boxes.

Failure decision tree

SymptomAction
Installer script fails to downloadCheck internet: ping -c 3 link.lumeo.com. Verify wget is installed: sudo apt install wget.
Gateway not appearing in consoleEnsure the default cloud gateway was deleted first (free tier limit). Re-run installer if needed.
Camera not discoveredClick "Discover" again. For USB cameras, check ls /dev/video*. For ONVIF, verify same network/subnet.
Pipeline deployment failsCheck gateway is online (green status). Verify camera is linked and producing preview snapshots.
WebRTC stream not loadingTry a different browser (Chrome recommended). Check Jetson network firewall rules.

Reference files

  • references/source.body.md — full original Seeed tutorial with screenshots, camera setup details, and pipeline configuration (reference only)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free