vss-deploy-dense-captioning

Use this skill when deploying standalone RT-VLM dense captioning or calling its REST API (uploads, captions, streams, chat-completions, Kafka). Not for VSS profile deploy or video-search ingestion.

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

Quality

Content

100%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A dense, actionable skill body with executable workflows, explicit validation checkpoints, and clean one-level-deep reference separation. It respects Claude's intelligence and routes complex detail to verified bundle files.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — no padding about what Docker, Kafka, SSE, or VLMs are — and every code block earns its place; only a couple of minor near-duplications (OpenAPI source-of-truth, alias clarification) could be trimmed, fitting the score-5 anchor best.	5 / 5
Actionability	It provides copy-paste-ready curl/bash with real endpoints, exact model-id retrieval via jq, env-var guards such as '${API_KEY:?}', and concrete RTSP validation, matching the score-5 anchor for fully executable guidance covering common cases.	5 / 5
Workflow Clarity	The standalone deployment flow is explicitly numbered with a 'Never skip the dry-run' checkpoint, preflights before pull, a smoke test after ready, and troubleshooting feedback loops (401 -> docker login retry; OOM -> switch variant / compose down); validation is present for this infra operation, matching the score-5 anchor.	5 / 5
Progressive Disclosure	SKILL.md acts as a clear overview with one-level-deep, well-signaled markdown links to real files (deploy-rt-vlm-service.md, api-surface-26.05.md, kafka-workflows.md), plus an upfront routing decision, matching the score-5 anchor.	5 / 5
	Total	20 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A precise, well-scoped description that states both capabilities and trigger conditions while explicitly disambiguating from sibling VSS skills. Minor room for additional natural synonyms, but otherwise strong.

Dimension	Reasoning	Score
Specificity	The description names the domain and enumerates multiple concrete actions — 'deploying standalone RT-VLM dense captioning' and 'uploads, captions, streams, chat-completions, Kafka' — giving comprehensive coverage, matching the score-5 anchor.	5 / 5
Completeness	It explicitly answers both what ('deploying standalone RT-VLM dense captioning or calling its REST API...') and when ('Use this skill when...') with concrete trigger phrases, matching the score-5 anchor.	5 / 5
Trigger Term Quality	Natural user terms are well covered ('deploying', 'dense captioning', 'REST API', 'uploads', 'captions', 'streams', 'chat-completions', 'Kafka'), but a few common variations a user might say (e.g. 'vision-language model', 'VLM') are missing, fitting the score-4 anchor better than 5.	4 / 5
Distinctiveness Conflict Risk	The explicit exclusion clause ('Not for VSS profile deploy or video-search ingestion') carves a clear niche with distinct triggers and minimal conflict risk with sibling deploy skills, matching the score-5 anchor.	5 / 5
	Total	19 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: NVIDIA/skills
Path: skills/vss-deploy-dense-captioning/SKILL.md
Commit: 994b870

Reviewed: ~15 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.