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 emit-reviewable-rrdgit 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/emit-reviewable-rrd)<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/emit-reviewable-rrd"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/emit-reviewable-rrd/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/emit-reviewable-rrd"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/emit-reviewable-rrd.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 | $0.00050 | $0.01651 |
| Opus 5 | $0.00025 | $0.00826 |
| Sonnet 5 | $0.00010 | $0.00330 |
| Haiku 4.5 | $0.00005 | $0.00165 |
Grade A, and why
emit-reviewable-rrd 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 3d 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Emit reviewable RRD artifacts
Build the recording from the run being reviewed. A stock example, generated placeholder, screenshot, or renamed JSON file is never evidence for the run.
Procedure
- Identify the actual stage outputs that contain the facts to visualize. Keep raw customer inputs private; extract only the metrics, events, frames, or trajectories needed for review.
- Choose timelines that preserve source semantics. Use
optimizer_stepfor training progress, real capture time for timestamped sensors, and an explicitly labelled dataset/frame index when capture time does not exist. - Choose stable entity paths before writing. For training, prefer grouped
entities such as
metrics/loss,metrics/learning_rate,throughput/global_samples_per_second,health/gradient_norm,checkpoint/materialized, andprovenance/run. - Set the Rerun recording id to the workflow run id. Record sanitized static provenance: producer, source revision, recipe/config identity, source artifact hashes, and factual limitations. Never embed credentials, signed URLs, customer payloads, hostnames, pod/node ids, or private infrastructure identifiers.
- Write the recording with
rerun-sdkand close its sink before inspection. Reuse an existing NPA Rerun converter or inspection helper when it matches the source; extend the producing workbench integration when it does not. - Put the file at a run-scoped private URI such as
s3://<bucket>/<workflow>/<run.id>/reports/<name>.rrd. Declare that exact URI in the producing state'soutputswith schemaapplication/vnd.rerun.rrd. Keep inputs and all companion artifacts under the same run prefix sonpa workbench workflow artifactsand artifact-first discovery can find them. - Fail the artifact stage if the required recording cannot be created, uploaded, or validated. Do not turn a mandatory RRD into a warning-only side effect.
Make the content reviewable
- For optimization, log factual loss, the exact applied learning-rate schedule,
interval timing/throughput, finite gradient or update-health diagnostics,
checkpoint events, and aggregate distributed/device health on the
optimizer_steptimeline. - Include before/after or held-out policy trajectories only when this run actually produced both sides with a valid alignment. Otherwise state the limitation in provenance and omit those entities.
- Use a blueprint when it materially improves the first view, but keep the underlying entities independently inspectable.
- Prefer a durable, deduplicated metric journal during long jobs and convert it deterministically after success. This makes resume factual without relying on unsupported append/recovery behavior for a partial RRD.
What ships with it
1 file 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.
- 3d ago Changed · +11 lines 5b62b56deef7
- 7d ago First seen · 132 lines · 50 tokens per session scan A 1db8264b5047
emit-reviewable-rrd is a skill published in the GitHub repository nebius/nebius-physical-ai (29 stars, last pushed today), licensed Apache-2.0. It adds 50 tokens to every session and 1,651 once invoked, about $0.0003 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 2026-09-05.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…