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 encordgit 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/encord)<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/encord"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/encord/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/encord"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/encord.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 | $0.00030 | $0.00797 |
| Opus 5.5 | $0.00012 | $0.00319 |
| Sonnet 5.5 | $0.00006 | $0.00159 |
| Haiku 4.5 | $0.00003 | $0.00080 |
Grade A, and why
encord 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 23d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Operate the Encord transport
Encord remains remote SaaS. NPA is a stateless CPU client that runs locally or in the default workflow pod. Do not deploy a service or container for this integration.
Preflight
Reference credentials by environment name or NPA credential configuration. Never embed values in a command, spec, example, receipt, or manifest.
npa workbench health preflight --checks encord
For workflow forwarding, use ENCORD_SSH_KEY_B64 because a local
ENCORD_SSH_KEY_FILE path cannot reach a pod.
Push
npa workbench encord push \
--input-path s3://<bucket>/raw-media/ \
--integration <encord-cloud-integration> \
--folder <encord-folder> \
--output-path s3://<bucket>/encord/push/push_receipt.json
Register mode retains S3 as the dataset of record. --transfer upload is an
explicit alternative that creates an Encord-managed copy with separate
retention and possible duplication. Never retry a registration failure as an
upload. If S3 has no full-object SHA-256, register reads every source byte
through the NPA client and requires s3:GetObject; it has no hashing opt-out.
Require exact source URI, complete object key or URL, namespaced metadata, stable item UUID, or a sidecar assertion. Never use a filename or basename as identity. Stop when exact signals conflict.
Pull
The default --label-export none avoids label initialization. Use
--label-export initialize only when the operator explicitly selects and
confirms that remote Encord label-state mutation. Retain the resulting manifest
as evidence. New pulls emit npa.encord.pull_manifest.v2, with exact transferred
source_size and separate provider_reported_size. Verification and rendering
continue to read v1 without rewriting its size fields.
Labeling demo
Partner workflows and their runbook live in workflows/partners/encord/.
encord-labeling-demo.yaml runs push → import-labels → project pull with label
initialization → media verification → render-labels. Import creates a new
ontology and project from a SHA-256-bound video plan (v1 boxes; v2 box/polygon
tracks and temporal classifications). It refuses
existing project titles and output receipts. Rendering compares the actual
exported row/object identities, frames, classes, coordinates, and classification
options/ranges, then produces
annotated MP4s from those exported labels. These are unreviewed programmatic
prelabels, not human-approved annotations. Keep that distinction in demos.
Select the new project and S3 targets before execution. Keep exact remote
identities and signed label-export media URLs in private evidence.
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.
- 23d ago Changed · +18 lines 89cedeaed3d9
- 27d ago First seen · 67 lines · 30 tokens per session scan A bdc8e3652650
encord is a skill published in the GitHub repository nebius/nebius-physical-ai (50 stars, last pushed today), licensed Apache-2.0. It adds 30 tokens to every session and 797 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 2026-09-25.
Other skills, from other repositories
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.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
google-cloud-solution-agentic-analytics-spark-knowledge-catalog
Discovers requirements and designs an end-to-end governed agentic analytics solution using Knowledge Catalog and Managed Service for Apache Spark (Lightning Engine). Use when designing data science and analytics workflows across structured and unstructured distributed data (including in S3, Azure Blob, AlloyDB, and…
training-check
Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
finetuning
Fine-tune models on Microsoft Foundry using SFT (supervised), DPO (preference), or RFT (reinforcement with graders). Covers dataset preparation, training job submission, deployment, and evaluation. USE FOR: fine-tune, SFT, DPO, RFT, training data, grader, distillation, fine-tuned model, training job, large file…