dynamo-router-starter

Start or patch Dynamo router modes and run router endpoint smoke checks. Use for round-robin, KV-aware, least-loaded, or device-aware routing setup; use recipe-runner for recipe deployment and troubleshoot for failure diagnosis.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

The canonical home for this skill is dynamo-router-starter in NVIDIA/skills

Quality

Content

82%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 executable commands, a clear numbered workflow, and verified one-level-deep bundle references. Tightening inline detail that overlaps router-modes.md and adding an explicit validate-pass checkpoint before comparison claims would push it higher.

Suggestions

Add an explicit gate after the smoke test (e.g. 'Only proceed to mode comparison once /v1/models and one chat completion succeed') to strengthen the validation checkpoint.

Trim env/CLI knobs duplicated inline (DYN_ROUTER_MODE, DYN_ROUTER_USE_KV_EVENTS) by pointing to references/router-modes.md, keeping SKILL.md as a true overview.

Label the one chat-completion smoke result as pass/fail explicitly in the Output Contract so limitations are unambiguous.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with well-scoped sections and minimal over-explanation of Dynamo internals; a few descriptive sentences in Purpose/Prerequisites could be trimmed, but every section earns its place.	4 / 5
Actionability	Fully executable copy-paste-ready commands (python3 -m dynamo.frontend --router-mode kv --http-port 8000, the smoke-test invocation, kubectl port-forward) plus concrete env patches and an invoked script with explicit arguments covering the common local and Kubernetes cases.	5 / 5
Workflow Clarity	Clear numbered sequence (baseline -> KV -> smoke test -> compare) with a smoke-test verification step and a troubleshooting table for error recovery; not a 5 because the destructive/batch-style comparison step lacks an explicit validate-before-claim checkpoint and the smoke test is stated as a 'must' rather than a gated pass/fail loop.	4 / 5
Progressive Disclosure	Well-organized SKILL.md overview with a single one-level-deep reference to references/router-modes.md (verified present) and scripts/check_router_health.py (verified present), clear sectioning, and an Available Scripts table; the inline env/CLI detail slightly duplicates router-modes.md, keeping it just short of a 5.	4 / 5
	Total	17 / 20 Passed

Description

86%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, specific description with concrete actions, explicit 'use for' trigger guidance, and clear boundary routing to sibling skills. Minor trigger-term synonym coverage could be broadened.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Start or patch Dynamo router modes', 'run router endpoint smoke checks', 'KV-aware, least-loaded, or device-aware routing setup') covering setup and testing; minor gaps in coverage (e.g. no explicit mention of approximate-mode fallback).	4 / 5
Completeness	Explicitly answers both 'what' ('Start or patch Dynamo router modes and run router endpoint smoke checks') and 'when' ('Use for round-robin, KV-aware, least-loaded, or device-aware routing setup') with concrete trigger phrases, and adds boundary guidance ('use recipe-runner for recipe deployment and troubleshoot for failure diagnosis').	5 / 5
Trigger Term Quality	Includes natural trigger terms users would say ('round-robin', 'KV-aware', 'least-loaded', 'device-aware routing', 'router endpoint smoke checks') with good keyword coverage; a few common synonyms or phrasings are missing.	4 / 5
Distinctiveness Conflict Risk	Clearly scoped to Dynamo router bring-up and smoke checks with a distinct niche, and explicitly disambiguates against adjacent skills (recipe-runner, troubleshoot), minimizing conflict risk.	5 / 5
	Total	18 / 20 Passed

Validation

87%

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 — 14 / 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

	Total	14 / 16 Passed

Repository: openai/plugins
Path: plugins/nvidia/skills/dynamo-router-starter/SKILL.md
Commit: 1dc1958

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.