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 nebius/nebius-physical-ai --skill open3dgit clone --depth 1 https://github.com/nebius/nebius-physical-aiWrote 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/nebius/nebius-physical-ai/open3d)<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/open3d"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/open3d/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/nebius/nebius-physical-ai/open3d"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/open3d.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 | $0.00040 | $0.03997 |
| Opus 5.5 | $0.00016 | $0.01599 |
| Sonnet 5.5 | $0.00008 | $0.00799 |
| Haiku 4.5 | $0.00004 | $0.00400 |
Grade A, and why
open3d 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 18d 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 — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Open3D point-cloud registration and reconstruction
Two scans of the same room do not share a coordinate frame. This tool recovers the
rigid transform between them with the real upstream pipeline —
registration_ransac_based_on_feature_matching over FPFH features, refined by
registration_icp — then optimizes the full pairwise pose graph with
global_optimization, fuses the result, and reconstructs a surface with
create_from_point_cloud_poisson. Nothing is recomputed or smoothed locally: every
fitness, RMSE and correspondence count is what Open3D's own RegistrationResult
returned.
CPU-only, and that is a design constraint rather than an unfinished port. Open3D's
pipelines.registration and geometry APIs have no CUDA path, so the image does not
derive from the CUDA base and no stage requests an accelerator. A stage that asked
for a GPU here would hold one it cannot use.
The image is 0.20.0-cpu-20260918, a validation candidate quarantined from public
publication until exact-image byte scans are accepted. Build from committed inputs
with npa/docker/workbench/open3d/build.sh.
Run it
npa workbench workflow validate-spec workflows/testing/open3d-registration.yaml
npa workbench workflow submit workflows/testing/open3d-registration.yaml \
--var bucket=<your-bucket>
The shipped spec starts from stage-demo, which downloads the upstream
open3d.data DemoICPPointClouds scans (release 20220301-data) — fragments of
the Redwood augmented ICL-NUIM benchmark (CC BY 3.0), depth frames rendered from
a modelled indoor scene with simulated sensor noise rather than real captures.
They carry the overlap and noise the pipeline expects, so the run needs no
operator capture to produce a real result, but nothing measured against them is
a statement about real-sensor accuracy. To register your own data, replace that
stage with prepare and point --input-path at a prefix holding at least two
.pcd/.ply files.
voxel_size is the single scale knob. The normal, FPFH and correspondence radii
are all multiples of it, exactly as the upstream Global-registration and
Multiway-registration tutorials derive them, so tuning one number rescales the
whole pipeline coherently. It is in the scans' own units; 0.05 suits
metre-scale indoor captures.
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.
- 18d ago First seen · 259 lines · 40 tokens per session scan A 6895c050bcf1
open3d is a skill published in the GitHub repository nebius/nebius-physical-ai (50 stars, last pushed today), licensed Apache-2.0. It adds 40 tokens to every session and 3,997 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 2026-10-04.
Other skills, from other repositories
instrument-data-to-allotrope
Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…
exploratory-data-analysis
Performs bounded, local exploratory analysis of explicitly supported scientific files. Supports redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…
matlab
Builds, reviews, migrates, and plans MATLAB or GNU Octave numerical workflows. Use for arrays, tabular/time data, tests, projects, graphics, MAT files, and explicit Python interoperability.
phylogenetics
Builds and analyzes phylogenetic trees using MAFFT multiple sequence alignment, IQ-TREE maximum likelihood with ModelFinder and branch support, and FastTree approximate inference. Uses ETE3 for tree summaries and visualization. Applies to homologous nucleotide or protein sequences, microbial gene trees, protein…
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
presenting-conference-talks
Generates conference presentation slides (Beamer LaTeX PDF and editable PPTX) from a compiled paper with speaker notes and talk script. Use when preparing oral talks, spotlight presentations, or invited talks for ML and systems conferences.