Borrowing it
Nothing to install: this file belongs to NVlabs/ASPIRE. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/NVlabs/ASPIRE/main/.agents/skills/cap-tool-usage/SKILL.mdgit clone --depth 1 https://github.com/NVlabs/ASPIREWrote 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/nvlabs/aspire/cap-tool-usage)<a href="https://agentmods.dev/skills/nvlabs/aspire/cap-tool-usage"><img src="https://agentmods.dev/badge/skills/nvlabs/aspire/cap-tool-usage/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/nvlabs/aspire/cap-tool-usage"><img src="https://agentmods.dev/badge/skills/nvlabs/aspire/cap-tool-usage.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 | $0.00033 | $0.01362 |
| Opus 5.5 | $0.00013 | $0.00545 |
| Sonnet 5.5 | $0.00007 | $0.00272 |
| Haiku 4.5 | $0.00003 | $0.00136 |
Grade A, and why
cap-tool-usage 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 21d 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CAP Tool Usage
Follow AGENTS.md. Use the native mcp__cap__* interface
attached to the ASPIRE CAP session. Keep that session as the only environment and
camera owner; do not create another robot or camera runtime during a physical task.
For a plainly short-horizon task, execute the shortest dependency-ordered MCP sequence immediately. Skip source inspection, a full planning report, and redundant checks.
ASPIRE action contract
Native MCP schemas are authoritative. The low-level action tools are:
| Tool | Target |
|---|---|
cap_move_to_pose |
target_pose, with arm="left" or "right" |
cap_move_synchronized |
targets={"left": pose, "right": pose} for jointly planned motion |
cap_move_to_joints |
target containing six joint angles in radians, with arm |
cap_set_gripper |
position from 0 closed to 1 open, with arm |
An absolute pose includes position: [x, y, z] in metres, frame, and exactly one
of rpy_deg: [roll, pitch, yaw] or quaternion_wxyz: [w, x, y, z]. RPY uses degrees
and fixed-axis XYZ: R = Rz(yaw) @ Ry(pitch) @ Rx(roll) in the declared frame.
Returned poses use WXYZ quaternions. Read measured arm poses with cap_get_robot_state.
For Cartesian or synchronized pose motion with cuRobo configured, select
strategy={"ik_solver":"mink","trajectory_planner":"curobo"}. Never disable
collision checking to force a failed plan through. Do not use direct joint motion
as implicit IK or to bypass a failed collision-aware Cartesian plan.
For a layer turn, derive successive absolute pose targets from the measured cube
axis and EEF pose. An offset EEF needs both its position and orientation rotated
about that axis. Verify contact and actual sticker motion from returned evidence.
Use separate cap_set_gripper calls for intended grip changes. Prefer synchronized
motion only when both targets are grounded and neither depends on the other's feedback.
Object-approach tool chain
Use the MCP schemas as authoritative. The expected discoverable tools are:
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.
- 21d ago First seen · 103 lines · 33 tokens per session scan A 04656145e0cd
cap-tool-usage is a skill published in the GitHub repository NVlabs/ASPIRE (959 stars, last pushed 22d ago), licensed Apache-2.0. It adds 33 tokens to every session and 1,362 once invoked, about $0.0001 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 2026-10-17.
cap-tool-usage compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| cap-tool-usageNVlabs/ASPIRE | Skill | 959 | 22d ago | Apache-2.0 | A | 33 | Claude Code, Codex |
| gke-compute-classesgoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 83 | Claude Code, Codex |
| jetson-diagnosticNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 30 | Claude Code, Codex |
| holoscan-install-wheelNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 34 | Claude Code, Codex |
| offensive-z-waveSnailSploit/Claude-Red | Skill | 10k | 25d ago | MIT | A | 113 | Claude Code, Codex |
| doca-socket-relayNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 236 | Claude Code, Codex |
| hsb-flashNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | B | 94 | Claude Code |