dynamo-troubleshoot

Diagnose failed or unhealthy Dynamo deployments. Use when pods, model-cache jobs, PVCs, workers, frontend/router health, endpoints, or benchmark jobs fail; use recipe-runner/router-starter before this for normal bring-up.

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Medium

Suggest reviewing before use

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.

The body is a well-structured, executable read-only troubleshooting runbook with clear sequencing, a feedback loop, and appropriately delegated detail. The only minor issue is slight redundancy between the Instructions and Examples sections.

Dimension	Reasoning	Score
Conciseness	The body is lean with exact kubectl/python3 commands and no padding explaining Kubernetes concepts, but the Examples section repeats the collect command already shown in Instructions and run_script is demonstrated twice, so minor trimming is possible per anchor 4.	4 / 5
Actionability	Commands are fully executable with real flags (--namespace, --deployment-name, --output-dir), the run_script protocol snippet is runnable, and curl examples in the reference cover common cases, matching anchor 5.	5 / 5
Workflow Clarity	A clear numbered sequence (collect, classify, debug top-down 1-8, fix one layer) includes an explicit feedback loop ("After each fix, rerun the relevant readiness check before moving deeper"), matching anchor 5.	5 / 5
Progressive Disclosure	SKILL.md is an overview that delegates per-bucket detail to references/failure-decision-tree.md (verified present) and the bundle script to scripts/, with references clearly signaled in a References section and Available Scripts table at one level of depth, matching anchor 5.	5 / 5
	Total	19 / 20 Passed

Description

87%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 clearly states purpose and triggers with concrete operator terminology and a useful boundary clause. Minor room to broaden the action vocabulary and add a few more colloquial failure terms.

Dimension	Reasoning	Score
Specificity	"Diagnose failed or unhealthy Dynamo deployments" names the domain and core action, and the trigger list enumerates concrete failure subjects (pods, PVCs, workers, frontend/router, benchmark jobs), but the action vocabulary is a single verb rather than several distinct actions, landing between anchors 4 and 5.	4 / 5
Completeness	It explicitly states what ("Diagnose failed or unhealthy Dynamo deployments") and when ("Use when pods, model-cache jobs, PVCs, workers, frontend/router health, endpoints, or benchmark jobs fail") with concrete trigger phrases, plus a boundary clause, matching anchor 5.	5 / 5
Trigger Term Quality	Natural operator terms are present (pods, PVCs, workers, frontend/router, endpoints, benchmark jobs, failed or unhealthy), but some common variations users say (e.g. CrashLoopBackOff, ImagePullBackOff, model download) are omitted, matching anchor 4.	4 / 5
Distinctiveness Conflict Risk	It carves a clear niche (Dynamo deployment troubleshooting) with distinct triggers and explicitly routes normal bring-up to recipe-runner/router-starter, minimizing conflict risk per anchor 5.	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: NVIDIA/skills
Path: skills/dynamo-troubleshoot/SKILL.md
Commit: 2ba0662

Reviewed: 5 hours 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.