nemotron-add-pattern

Add a cross-cutting decision pattern under src/nemotron/steps/patterns/. Use when a recurring ML decision (tokenizer lock, eval bookends, LoRA-on-small-data, etc.) must be encoded so other skills can fire it during planning.

Quality

85%

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

88%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 lean, highly actionable workflow with explicit sequencing, validation checkpoints, and error-recovery loops. Its only weak spot is mild command repetition across sections and a calibration-examples block that could be externalized.

Suggestions

Centralize the catalog/test commands in one place and reference them from other sections to reduce repetition of 'uv run python src/nemotron/steps/index.py' and 'uv run pytest tests/steps -q'.

Consider moving the Calibration Examples into a separate references file (e.g. references/calibration-examples.md) to improve progressive disclosure and shorten the main SKILL.md.

Tighten the Boundaries Do/Don't lists, which restate rules already covered in the Workflow and Generation rules sections.

Dimension	Reasoning	Score
Conciseness	The body is checklist-first and assumes Claude's competence without explaining known concepts, but the catalog/test commands ('uv run python src/nemotron/steps/index.py', 'uv run pytest tests/steps -q') and 'regenerate PATTERNS.md' recur across Tone, Generate, Validate, Summarize, and Boundaries, which is minor trimmable repetition.	4 / 5
Actionability	Guidance is fully executable: copy-paste commands, exact file paths, required frontmatter fields, required body sections, and two calibration examples covering the common cases, matching the top anchor.	5 / 5
Workflow Clarity	Four phases are rigidly sequenced (Orient, Generate, Validate, Summarize) with an explicit validation checklist and a feedback loop ('If validation fails: fix, re-run, do not present until tests pass') plus a 'When Stuck' escalation, satisfying the top anchor for batch file-generation work.	5 / 5
Progressive Disclosure	Sections are well-organized and references (to live repo files like small-dataset-lora.md and PATTERNS.md) are clearly signaled, but no bundle files exist and the Calibration Examples block is the kind of supplementary material that could be split into a separate reference file, leaving it just shy of the top anchor.	4 / 5
	Total	18 / 20 Passed

Description

82%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 concise, third-person, and explicitly covers both what the skill does and when to use it with concrete domain triggers. Its only limitation is that it names a single action rather than a comprehensive set of capabilities.

Dimension	Reasoning	Score
Specificity	It names the domain ('cross-cutting decision pattern') and one concrete action ('Add ... under src/nemotron/steps/patterns/') with a specific target path, but does not list multiple distinct actions, so it sits at the '1-2 concrete actions' anchor rather than the multi-action anchor above.	3 / 5
Completeness	It explicitly answers what ('Add a cross-cutting decision pattern under src/nemotron/steps/patterns/') and when ('Use when a recurring ML decision ... must be encoded so other skills can fire it during planning') with concrete trigger examples, matching the top anchor.	5 / 5
Trigger Term Quality	Natural practitioner phrases like 'tokenizer lock', 'eval bookends', 'LoRA-on-small-data', and 'recurring ML decision ... encoded' give good keyword coverage; it stops short of comprehensive synonyms/extensions, fitting the 'good coverage, a few natural terms missing' anchor.	4 / 5
Distinctiveness Conflict Risk	The niche is tightly scoped to a specific repo path and ML-pattern-encoding task with concrete example triggers, giving a clear niche with minimal conflict risk against other skills.	5 / 5
	Total	17 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: NVIDIA-NeMo/Nemotron
Path: skills/nemotron-add-pattern/SKILL.md
Commit: 1a4c3fa

Reviewed: 4 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.