Claw3D is a self-hosted 3D virtual office where AI agents can collaborate, review code, run standups, execute tasks, and display their work in real time. It is for people running AI teams on their own infrastructure, with support for several agent runtimes. The catalogue skills and instruction support working with this agent workspace.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add iamlukethedev/Claw3D --skill soundclawgit clone --depth 1 https://github.com/iamlukethedev/Claw3DWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/iamlukethedev/claw3d/soundclaw)<a href="https://agentmods.dev/skills/iamlukethedev/claw3d/soundclaw"><img src="https://agentmods.dev/badge/skills/iamlukethedev/claw3d/soundclaw/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/iamlukethedev/claw3d/soundclaw"><img src="https://agentmods.dev/badge/skills/iamlukethedev/claw3d/soundclaw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 57 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00018 | $0.00787 |
| Opus 5 | $0.00009 | $0.00394 |
| Sonnet 5 | $0.00004 | $0.00157 |
| Haiku 4.5 | $0.00002 | $0.00079 |
Grade A, and why
soundclaw scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- soundhermes — 88% identical, 38 lines differ
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SOUNDCLAW
Use this skill when the user wants an agent to search for music, play a song or playlist, control Spotify playback, or send back a shareable Spotify link on the same channel the request came from.
Trigger
{
"activation": {
"anyPhrases": [
"spotify",
"play a song",
"play this song",
"play music",
"play a playlist",
"find a song",
"queue this song",
"music link"
]
},
"movement": {
"target": "jukebox",
"skipIfAlreadyThere": true
}
}
When this skill is activated, the agent should walk to the office jukebox before handling the request.
- Treat requests from Telegram or any other external surface as valid triggers when they ask for Spotify playback, search, queueing, or music-link sharing.
- The physical behavior for this skill is: go to the jukebox, perform the music-selection workflow, then report the result.
- If the agent is already at the jukebox, continue without adding extra movement narration.
Channel behavior
- Reply on the same active channel or session that received the request.
- If playback cannot start but a matching track, album, or playlist is found, send back the best Spotify link instead of failing silently.
- If multiple matches are plausible, ask a clarifying question instead of guessing.
OpenClaw Gateway Skill Contract
This section is for developers implementing the backend skill handler in OpenClaw. The Claw3D UI handles authentication via Spotify PKCE OAuth in the browser. The gateway skill handles agent-driven requests via the
soundclaw.*RPC namespace.
Authentication model
The user authenticates directly in the browser (PKCE, no secret required).
The access token is stored in browser localStorage under the key soundclaw_token.
For agent-driven playback (e.g. "play Jazz for me"), the gateway skill should either:
- Use a server-side Spotify app token (Client Credentials) for search-only actions, or
- Instruct the agent to tell the user to use the jukebox panel for actual playback
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 93 lines · 18 tokens per session scan A 60c9638460fd
soundclaw is a skill published in the GitHub repository iamlukethedev/Claw3D (2,206 stars, last pushed 18d ago), licensed MIT. It adds 18 tokens to every session and 787 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
soundhermes
Control Spotify playback, search music, and return shareable music links.
task-manager
Capture actionable user requests as persistent tasks, update task status as work progresses, and keep a shared task store in sync. Use when a user asks an agent to do work, check progress, block a task, complete a task, or manage the Kanban board.
todo
Maintain a shared workspace TODO list with blocked tasks.
canvas
Create, inspect, and update Obsidian JSON Canvas boards with text, file, link, group, and edge nodes. Use for canvas status, canvas lists, visual maps, zones, spatial layouts, adding vault notes or media to a .canvas file, and requests such as create canvas, add to canvas, or put this on the canvas.
youtube-shorts-generator
Generate viral 9:16 YouTube Shorts (or TikTok/Reels clips) from a long-form YouTube URL or local video. Triggers on requests like "make shorts from this video", "extract viral clips from this YouTube link", "auto-clip this podcast", "find the best moments and crop vertical". Pipeline downloads the source, transcribes…
ax-java-audio
Use when writing Java code with dev.axllm:ax for audio input/output, OpenAI Responses audio mapping, realtime event folding, and generated package audio examples.