CtrlK
BlogDocsLog inGet started
Tessl Logo

nemotron-add-step

Add a new step under src/nemotron/steps/<category>/<step_id>/ — manifest (step.toml), runner glue, configs, and per-step README.md. Use when extending the catalog so /nemotron-customize can route to it.

75

Quality

92%

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

100%

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

An exemplar instruction-only skill body: lean, actionable, with a clearly sequenced workflow and explicit validation feedback loops. No bundle files are needed and none are referenced, so progressive disclosure is satisfied by clean in-file organization.

DimensionReasoningScore

Conciseness

Checklist-first, bullet-driven content with specific paths and no conceptual padding; assumes Claude's competence ('Concise. Checklist-first.') and every section earns its place.

3 / 3

Actionability

Concrete executable commands ('uv run pytest tests/steps -q', 'uv run python src/nemotron/steps/index.py') and exact schema fields ('[step].id', '[[strategies]]', '[[errors]]') make the guidance copy-paste ready.

3 / 3

Workflow Clarity

Four explicitly ordered phases (Orient → Generate → Validate → Summarize) with a validation checkpoint, fix-and-re-run feedback loop, and 'Do not present the result until both pass' gate.

3 / 3

Progressive Disclosure

No bundle files are present; the skill is self-contained and well-organized into clearly signaled sections (Tone, Workflow, Boundaries, When Stuck), satisfying the simple-skill exception for progressive disclosure.

3 / 3

Total

12

/

12

Passed

Description

85%

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, specific description with a clear what-and-when structure and low conflict risk. The main weakness is trigger phrasing that leans on internal jargon ('extending the catalog') rather than terms a contributor would naturally say.

Suggestions

Add natural-language trigger variations a contributor might actually say, e.g. 'Use when adding a new step or step type to the nemotron-customize library'.

Reconcile the description with the body: it promises 'per-step README.md' but the workflow generates config/default.yaml, config/tiny.yaml, and a regenerated STEPS.md instead — align the listed artifacts.

Consider adding '.toml manifest' or 'step manifest' as explicit trigger keywords to broaden natural-term coverage.

DimensionReasoningScore

Specificity

Quotes concrete actions: 'manifest (step.toml), runner glue, configs, and per-step README.md' — multiple specific artifacts named, matching the score-3 anchor.

3 / 3

Completeness

Explicitly answers both what ('Add a new step under ... manifest, runner glue, configs') and when ('Use when extending the catalog'), matching the score-3 anchor with an explicit trigger clause.

3 / 3

Trigger Term Quality

Trigger 'Use when extending the catalog so /nemotron-customize can route to it' is relevant but domain-jargon-heavy; it lacks common natural variations like 'add a step' or 'new step type'.

2 / 3

Distinctiveness Conflict Risk

Niche is narrow and specific to the nemotron-customize step catalog with distinct triggers; unlikely to fire for unrelated skills.

3 / 3

Total

11

/

12

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-NeMo/Nemotron
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.