nemotron-retrieval-recipes

Use when planning, debugging, tuning, evaluating, exporting, or deploying public Nemotron `embed`/`rerank` retrieval recipes.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

92%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 well-structured, highly actionable skill body with a sequenced, validated workflow and clean progressive disclosure to real reference files. The main nit is some prerequisite content duplication between the Prerequisites section and the Safe Workflow.

Suggestions

De-duplicate the prerequisite list: keep it in Prerequisites and have Safe Workflow step 3 reference it rather than restating Stage 0/1-3/4/5 prerequisites inline.

The Limitations already warn that bundled references are condensed snapshots; consider moving the per-stage prerequisite detail that lives in both sections into references/embed.md and references/rerank.md to shorten the body further.

Dimension	Reasoning	Score
Conciseness	Dense and largely token-efficient with concrete commands rather than concept explanation, but prerequisite material repeats across the Prerequisites section and Safe Workflow step 3 and could be tightened.	4 / 5
Actionability	Copy-paste-ready commands ("uv run nemotron <family> <stage> -c <profile> -d"), real config paths, dotlist-override guidance, and explicit preview-vs-execution separation make it fully executable.	5 / 5
Workflow Clarity	The Safe Workflow is a clearly sequenced checklist with cheap-check-first ordering, prerequisite verification, and validation via dry-runs before any long-running or batch execution.	5 / 5
Progressive Disclosure	The body is a concise overview that clearly signals one-level-deep references (references/embed.md, rerank.md, evaluation.md, remote.md), all of which exist, with navigation in both Instructions and a References section.	5 / 5
	Total	19 / 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 strong, explicit trigger description naming a clear product niche and enumerating concrete lifecycle actions. Minor room to add synonyms or file-extension/CLI-style trigger terms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "planning, debugging, tuning, evaluating, exporting, or deploying" — tied to the named Nemotron embed/rerank recipe domain, giving comprehensive lifecycle coverage.	5 / 5
Completeness	Explicitly answers both: "what" (public Nemotron embed/rerank retrieval recipes) and "when" (the Use when clause enumerates trigger tasks).	5 / 5
Trigger Term Quality	Natural verbs users say (planning, debugging, tuning, evaluating, deploying) plus product terms (Nemotron embed/rerank) give good coverage, but a few synonyms or file/CLI forms are missing.	4 / 5
Distinctiveness Conflict Risk	The "public Nemotron embed/rerank retrieval recipes" niche is highly specific with distinct triggers and minimal overlap with other skills.	5 / 5
	Total	19 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: NVIDIA-NeMo/Nemotron
Path: skills/nemotron-retrieval-recipes/SKILL.md
Commit: 907d408

Reviewed: 6 days 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.