agentleFS
facebookincubatorfacebookincubator/muse-gadget-sdk/skills/gadget-pixoo-64/SKILL.md

gadget-pixoo-64

Skill2.8k starsChanged 26 days agoApache-2.0View on GitHub

Drive a Divoom Pixoo-64 LED pixel display (64x64) on the local network through Muse Home Link — read its state, push still frames and text — using the device's local HTTP API.

Trending

More talk lately.thin dataabout the whole repository

Dec 2025: 0 mentionsJan 2026: 0 mentionsFeb 2026: 0 mentionsMar 2026: 0 mentionsApr 2026: 0 mentionsMay 2026: 0 mentionsJun 2026: 0 mentionsJul 2026: 0 mentionsAug 2026: 0 mentionsSep 2026: 0 mentionsOct 2026: 13 mentionsNov 2026: 1 mention
Dec 2025Jun 2026Nov 2026
Mentions a month of the facebookincubator/muse-gadget-sdk repository that our weekly crawl found: only 14 in all, so this is a hint, not a trend. There is no adoption data for this file yet.

Who's talking about facebookincubator/muse-gadget-sdk

The biggest 5 of 14 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 14

Talked about on GitHub, Bluesky and Hacker News (14 times).

On GitHub · 11

6 more on GitHub

On Hacker News · 1

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 0 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read and follow the Home Link networking and safety rules in ~/docs/devices/home_link.md.

    line 19
  2. Very specific:

    Every command is a JSON POST to a single endpoint on port 80:

    line 23
  3. Very specific:

    Every response is JSON and includes error_code — 0 means success.

    line 32
  4. Very specific:

    Always send Draw/ResetHttpGifId before a frame or frame sequence — without it, the device silently ignores the frames.

    line 62
  5. Very specific:

    Use only the commands above.

    line 87

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

Divoom Pixoo-64

The Pixoo-64 is a 64×64 LED pixel display with an unauthenticated local HTTP API. It works with Muse Home Link when it is on the same subnet as the Link: find it with Link discovery, then send its API calls through the Link as documented HTTP requests. No cloud account or Divoom app is needed for the local API.

Verified working with Home Link on 2026-11-12 (a Channel/GetIndex call through the Link returned the device's current channel).

Prerequisites

Read and follow the Home Link networking and safety rules in ~/docs/devices/home_link.md.

The API

Every command is a JSON POST to a single endpoint on port 80:

POST http://<device-ip>/post
Content-Type: application/json

{"Command": "<command>", ...parameters}

Every response is JSON and includes error_code — 0 means success. Check it on every call; the device accepts some malformed requests silently.

Read state (safe, changes nothing)

  • {"Command":"Channel/GetIndex"} — returns SelectIndex, the current channel. Channel 3 is the custom channel, where frames you push are shown.
  • {"Command":"Channel/GetAllConf"} — returns the device's channel configuration. The field set varies by firmware; use Channel/GetIndex when you only need the current channel.

Show something

Note the current channel (Channel/GetIndex) so you can switch back when asked, then switch to the custom channel before pushing frames:

{"Command":"Channel/SetIndex","SelectIndex":3}

Push one still frame. PicData is raw RGB pixel bytes, base64-encoded — 64 × 64 × 3 = 12,288 bytes for a full frame, row by row, one byte each for R, G, B per pixel. It is not a PNG or GIF file; render the frame to raw pixels first.

{"Command":"Draw/ResetHttpGifId"}
{"Command":"Draw/SendHttpGif","PicNum":1,"PicWidth":64,"PicOffset":0,"PicID":0,"PicSpeed":100,"PicData":"<base64 raw RGB>"}

Always send Draw/ResetHttpGifId before a frame or frame sequence — without it, the device silently ignores the frames.

Draw text over a frame you have pushed:

{"Command":"Draw/SendHttpText","TextId":1,"x":0,"y":0,"dir":0,"font":8,"TextWidth":64,"TextHeight":16,"speed":50,"color":"#FFFFFF","TextString":"HELLO"}

Limits

  • Stills first. Pushing multi-frame GIF animations can crash and reboot the device (observed with a 16-frame push on current firmware); community experience caps animations at roughly 40 frames even when they work. Prefer single still frames. To animate, re-push stills yourself at a human pace instead of handing the device an animation.
  • Pace your requests. The device handles rapid-fire requests badly — under ~100 ms spacing is unreliable. Keep at least ~350 ms between POSTs in a sequence.
  • Draw/CommandList cannot batch animation frames. Send one Draw/SendHttpGif per frame.
  • Text only overlays frames sent via Draw/SendHttpGif — it does not draw over gallery or SD-card content.
  • The local API has no authentication. Anyone on the subnet can drive the display. Keep the device on a trusted LAN segment and never port-forward it.
  • Use only the commands above. No reboot, settings, firmware or URL/SD-card playback commands (such as Device/PlayTFGif).

Sources

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.