CtrlK
BlogDocsLog inGet started
Tessl Logo

vss-search-archive

Use this skill to run top-level VSS fusion search on archived video, or to ingest video files / RTSP streams for search. Do NOT use for ad-hoc visual Q&A (use vss-ask-video), live captioning (use vss-deploy-dense-captioning), or video summarization and reports (use vss-summarize-video).

67

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

The body is a strong, executable operational runbook with copy-paste bash, a numbered mandatory workflow, and well-signaled reference files. Its weaknesses are mild redundancy and the destructive delete path lacking an explicit post-delete verification checkpoint.

Suggestions

Add an inline validation step after the DELETE calls (e.g. re-list sources via vss-manage-video-io-storage and confirm the source is gone) to lift workflow_clarity above the destructive-operation cap of 3.

Consolidate the repeated 'do not use bare-VIOS PUT / no embeddings' guidance into a single Gotchas entry and reference it, to tighten conciseness toward 4.

Consider moving the full three-step chunked-upload recipe into references/ and keeping a short summary inline, to push progressive_disclosure toward 5.

DimensionReasoningScore

Conciseness

The body is mostly efficient and operational, but repeats the bare-VIOS-vs-agent-backend warning across several sections and inlines the full three-step upload flow that partly overlaps the canonical recipe, fitting 'mostly efficient but could be tightened' rather than the minor-trim 4.

3 / 5

Actionability

It provides fully executable, copy-paste-ready bash for chunked upload, RTSP add, delete, and REST search, with jq parsing and an explicit sensorId failure check, covering the common cases.

5 / 5

Workflow Clarity

Although the mandatory workflow is well-sequenced with HARD STOPs, the destructive delete operation and batch ingestion lack an inline validation checkpoint confirming the source is gone / embeddings removed before declaring success, so the missing-validation cap holds it at 3.

3 / 5

Progressive Disclosure

Structure is good with clearly signaled one-level-deep references to real files (references/troubleshooting.md, references/discovery_modes.md) for bulk detail, but a fair amount of inline content (the full upload recipe, control-knob table) keeps it just below the ideal 5.

4 / 5

Total

15

/

20

Passed

Description

80%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.

The description is strong: it states concrete capabilities, provides explicit 'Use this skill to...' trigger guidance, and sharply disambiguates from sibling skills. Its main weakness is trigger-term quality, relying on operational phrasing rather than the natural keyword variations users would actually say.

Suggestions

Add natural-language trigger phrases users would say (e.g. 'search archived video for forklifts', 'ingest an RTSP camera for search', 'remove a video and its embeddings') to lift trigger_term_quality above 3.

Consider naming the underlying modality (Cosmos Embed1 embeddings / Elasticsearch) only if it surfaces a user-facing trigger term; otherwise keep it out of the description to avoid jargon.

DimensionReasoningScore

Specificity

Lists several concrete actions ('run top-level VSS fusion search on archived video', 'ingest video files / RTSP streams for search') but does not enumerate every sub-operation, fitting the 'several specific actions; minor gaps' anchor rather than the comprehensive 5.

4 / 5

Completeness

It explicitly answers both what ('run top-level VSS fusion search... ingest video files / RTSP streams') and when ('Use this skill to...'), with concrete trigger phrasing, matching the anchor that requires both what AND when with concrete triggers.

5 / 5

Trigger Term Quality

It states the capability and explicit 'Do NOT use' routing but uses paraphrased operational language rather than the natural keyword variations a user would say, fitting 'some relevant keywords but missing common variations or synonyms' rather than the fuller coverage of 4.

3 / 5

Distinctiveness Conflict Risk

It scopes to a clear niche (archived video search + ingest for search) and explicitly disambiguates from three sibling skills (vss-ask-video, vss-deploy-dense-captioning, vss-summarize-video), giving a distinct trigger set with minimal conflict risk.

5 / 5

Total

17

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
NVIDIA/skills
Reviewed

Table of Contents

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.