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 graph-robots/open-robot-skills --skill perceiving-objects-oneshotgit clone --depth 1 https://github.com/graph-robots/open-robot-skillsWrote 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/graph-robots/open-robot-skills/perceiving-objects-oneshot)<a href="https://agentmods.dev/skills/graph-robots/open-robot-skills/perceiving-objects-oneshot"><img src="https://agentmods.dev/badge/skills/graph-robots/open-robot-skills/perceiving-objects-oneshot.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00149 | $0.01539 |
| Opus 5 | $0.00075 | $0.00770 |
| Sonnet 5 | $0.00030 | $0.00308 |
| Haiku 4.5 | $0.00015 | $0.00154 |
Grade A, and why
perceiving-objects-oneshot 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 8d 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 — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
perceiving-objects-oneshot
Single VLM call over a set-of-marks overlay. Pipeline:
observe → perceive → filter_obb
perceive runs:
grounding-dino.detectwith a broadobject.text prompt- One
vlm.queryshowing the image with letter-labeled boxes: "Which letter is the ? Reply with one letter or 'none'." - On
none: emitfound: Falseso the subgraph exitsnot_found. On a letter:sam3.segment_boxon the chosen box,geometry.mask_to_world_pointsfor the cloud.
When to use
- Clean-all-items / multi-item loops where the cycle needs a clean
"no match" signal to terminate via
target.not_found → done. - Tasks where the target description is generic ("any item on the floor", "the next remaining grocery item") rather than a specific scene-spec id.
- Uncluttered scenes with distinct, reasonably sized targets where the set-of-marks letter pick is reliable.
When NOT to use
- Small / cluttered targets (< 40 px wide) — prefer
perceiving-objectswhose pairwise crop tournament is far more reliable in that regime.
Recommended subgraph state flow
3 states: observe → perceive → filter_obb (mirrors perceiving-objects).
State details:
About
object_namebelow: it is a literal Python string — the natural noun phrase for the object you are perceiving, drawn from this subgraph's description (e.g."alphabet soup","basket","any grocery item on the floor"). It is a constant per subgraph instance, NOT a binding. DO NOT writeRef("in.object_name")or any otherRef(...); the coordinator does not declareobject_nameas a subgraph input. Write the string directly, e.g."object_name": "any grocery item on the floor". The same rule applies toobject_descriptionif you set it.
observe—type: tool,tool: "robot.get_observation",inputs: {}. Connector tool; flat name only.perceive—type: script, filescripts/<sg>/perceive_simple.pyfrom this bundle. Inputs:cameras=Ref("observe.cameras"),object_name="<noun phrase from the subgraph description>", plus any optional literals (object_description,dino_prompt). Returns{found, cloud, mask, score}. When the VLM picks "none" or DINO emits no detections,foundisFalseand the downstreamfilter_obbstep then raises (empty cloud) — caught by the subgraph'son_error: "not_found"exit.filter_obb—type: tool,tool: "geometry.filter_and_compute_obb",inputs={"points": Ref("perceive.cloud")}. Returns{"obb": <OrientedBoundingBox>}.
What ships with it
5 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.
- 8d ago First seen · 150 lines · 149 tokens per session scan A 1862a6bb19d7
perceiving-objects-oneshot is a skill published in the GitHub repository graph-robots/open-robot-skills (39 stars, last pushed yesterday), licensed Apache-2.0. It adds 149 tokens to every session and 1,539 once invoked, about $0.0007 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
mjlab-to-mjswan
Port an mjlab task to mjswan: turn a local or remote GitHub repo that registers mjlab tasks into a browser app. Use when the user wants to run, visualize, or deploy an mjlab task (theirs or a third party's) in the browser with mjswan.
gap
Program robots with GaP (graph-as-policy) — compile natural-language tasks into typed, verified robot skill graphs and run them on simulators or real robots. Use when the user mentions GaP or graph-as-policy, robot manipulation, robot skills, robot tools or capabilities, skill registries, open-robot-skills, LIBERO or…
habitat-gs-train
Train and evaluate a navigation policy in the habitat-gs simulator. Covers the full generate-episodes → train → evaluate flow for PointNav / ImageNav / ObjectNav (Habitat-Lab + DDPPO reinforcement learning) and for Vision-and-Language Navigation (StreamVLN, Uni-NaVid). Use when the user wants to train, fine-tune…
authoring-pipelines
Use when the user wants a bagel data pipeline — reducing a log to windows around events ("keep 30s around every hard brake"), recurring snippets/GIFs/exports, or any tasks-and-gates automation over robot data.
agent-platform-rag-engine-management
Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.