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 robium-ai/robium --skill gemini-roboticsgit clone --depth 1 https://github.com/robium-ai/robiumWrote 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/robium-ai/robium/gemini-robotics)<a href="https://agentmods.dev/skills/robium-ai/robium/gemini-robotics"><img src="https://agentmods.dev/badge/skills/robium-ai/robium/gemini-robotics/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/robium-ai/robium/gemini-robotics"><img src="https://agentmods.dev/badge/skills/robium-ai/robium/gemini-robotics.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00039 | $0.01127 |
| Opus 5.5 | $0.00016 | $0.00451 |
| Sonnet 5.5 | $0.00008 | $0.00225 |
| Haiku 4.5 | $0.00004 | $0.00113 |
Grade A, and why
gemini-robotics 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gemini Robotics
Keep Gemini on the perception and planning side of an actuator boundary. The model may choose a capability; deterministic robot software validates and executes it.
Choose the endpoint from the interaction
- For a new robot assistant or first simulation demo, read architect before building a model/robot integration. It discovers the saved apps checkout and checks compatible examples. If a baseline was already selected, continue here; existing integration fixes and API explanations do not need onboarding.
- Use
gemini-robotics-er-2-streaming-previewfor a stateful Live API session that receives text, JPEG frames, or audio and orchestrates robot tools with low latency. - Use
gemini-robotics-er-2-previewfor discrete embodied-reasoning requests such as spatial analysis or offline video work. The standard endpoint does not provide the Live API. - Treat both model IDs and feature support as preview surfaces. Re-check the current model overview before changing dependencies or deployment assumptions.
Make the stream an observe-act-observe loop
- Keep one
client.aio.live.connectsession open for the task and run a receive loop that handles both model content and tool calls. - Serialize user turns and model-facing heartbeats around unresolved turns and blocking tools. A text heartbeat is a new reasoning input, not a transport keepalive, and can interrupt an action as barge-in.
- Declare physical actions with
behavior: BLOCKING. Execute each call through the robot adapter, then manually return aFunctionResponsewith the call ID, name, and structured result usingsend_tool_response. - Stream raw 16-bit, 16 kHz, little-endian PCM for speech input and explicitly
end finite audio with
audio_stream_end=True. Send JPEG camera frames at no more than the endpoint's current one-frame-per-second limit. - A camera frame alone updates context but does not trigger reasoning. Pair it with user audio/text, or use an intentional heartbeat prompt. Heartbeats are turns and can interrupt generation.
- When a tool exists specifically to observe the world, attach its fresh image
to that call's
FunctionResponsewhen the SDK supports inline media. This binds the evidence to the requesting call more deterministically than placing an unrelated realtime frame immediately before the response. - The streaming endpoint returns text, not synthesized audio. Route speech through an independently replaceable TTS adapter or expose speaking as a bounded tool.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago Changed · +5 lines · +10 tokens per session 5274f9e4a396
- 1mo ago First seen · 90 lines · 29 tokens per session scan A b30d0543246c
gemini-robotics is a skill published in the GitHub repository robium-ai/robium (99 stars, last pushed 16d ago), licensed MIT. It adds 39 tokens to every session and 1,127 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-18.
Other skills, from other repositories
ros2-microros
Skill "ros2-microros" from Leehyunbin0131/claude-ros2-skills, covering micro-ros instructions (ubuntu 24.04 lts & ros 2 jazzy), 1. architecture, 2. documentation entry points, 3. key concepts & patterns and a. embedded client node setup (rclc in c).
ros2-troubleshooting
Diagnose ROS 2 Jazzy faults that reading the code cannot settle: QoS mismatch, sensor-mount and frame errors (REP 103/105), and odometry calibrated against CAD instead of the floor. Ships four runnable pass/fail checks.
attach-end-effector
Mount an end-effector (parallel-jaw gripper, dexterous hand, suction tool, etc.) on a robot arm by attaching its MJCF to the arm's attachmentsite via MuJoCo's MjSpec API. Use when combining an arm MJCF and an end-effector MJCF into a single combined model. Handles known gotchas the underlying script doesn't: mesh…
urdf-to-mjcf
Use when bringing a URDF robot, hand, or CAD-generated model into MuJoCo MJCF and no hand-tuned upstream MJCF exists.
mjcf-to-urdf
Convert an MJCF (MuJoCo's XML format) to URDF for downstream consumers that don't speak MJCF (RViz, pinocchio, ROS-based motion planning, etc.). The conversion is lossy — constraints, position actuators, contact excludes, and keyframes are all dropped. Use this skill ONLY when a URDF-only consumer needs the kinematic…
gazebo-mcp
Gazebo / gz-sim worlds, models, poses (mock + live bridge). CLI gazebo-mcp + MCP stdio serve. Use when the user mentions gazebo-mcp, /gazebo-mcp, or related domain work. One-command Grok install from GitHub.