CtrlK
BlogDocsLog inGet started
Tessl Logo

ov-debug

Troubleshooting all sorts of failures, crashes, exceptions and errors using debug capabilities. Analyze accuracy, performance, model compilation, or memory issues. Dump tensors and intermediate blobs. Serialize and visualize IRs, execution graphs. Enable verbose, logging. Profile execution. Compare layer outputs. Inspect, trace or dump transformations. Identify executed operations, nodes, primitives, kernels.

60

Quality

71%

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

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/ov-debug/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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 concise, well-structured router whose body defers all substantive guidance to component reference files — but those files are missing, breaking progressive disclosure and leaving the body's actionability incomplete.

Suggestions

Ship the referenced component files (components/debug-intel-cpu-plugin.md, debug-intel-gpu-plugin.md, debug-transformations.md) so the routing table's links resolve; without them the skill is non-functional.

Add a minimal inline fallback (one or two concrete debug commands or env-var examples per component) so the body is actionable even before a reference is loaded.

Add a brief verification checkpoint in the Steps (e.g., confirm the build flag is present in CMakeCache.txt before proceeding) to strengthen workflow clarity.

DimensionReasoningScore

Conciseness

Very lean — a two-line prerequisites block, a routing table, and three steps — with no padding and no explanation of concepts Claude already knows; every token earns its place.

5 / 5

Actionability

The routing table gives concrete symptom-to-file mapping and a concrete build flag, but the substantive debugging actions are entirely deferred to reference files, leaving the key executable details missing from the body itself.

3 / 5

Workflow Clarity

A clear three-step sequence (match symptom -> load reference -> follow instructions) appropriate to this simple routing skill, with no checkpoints but no destructive operation requiring validation either.

4 / 5

Progressive Disclosure

The references are clearly signaled and one level deep in the table, but the referenced files (components/debug-intel-cpu-plugin.md, debug-intel-gpu-plugin.md, debug-transformations.md) and the components/ directory do not exist, so the disclosure path is a dead end and navigation is broken.

2 / 5

Total

14

/

20

Passed

Description

75%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 richly specific, domain-distinct description that enumerates many concrete debug capabilities, but it omits any explicit 'Use when...' trigger guidance, capping its completeness.

Suggestions

Append an explicit 'Use when...' clause naming the natural trigger phrases (e.g., 'Use when debugging OpenVINO inference failures, wrong results, slow inference or compilation, or unexpected model behavior').

Soften the wall of technical jargon by leading with the user-facing symptom terms (crashes, wrong results, slow inference, memory errors) before the lower-level capabilities.

Add a few natural synonyms / file extensions users might mention to lift trigger-term coverage from good to comprehensive.

DimensionReasoningScore

Specificity

Lists many concrete debug actions — 'Dump tensors and intermediate blobs', 'Serialize and visualize IRs, execution graphs', 'Profile execution', 'Compare layer outputs', 'Inspect, trace or dump transformations' — giving comprehensive coverage rather than just several.

5 / 5

Completeness

The 'what' is clear and extensive, but there is no 'Use when...' clause or equivalent explicit trigger guidance, so per the cap rule completeness cannot exceed 3.

3 / 5

Trigger Term Quality

Good coverage of natural symptom terms ('failures, crashes, exceptions and errors', 'performance, model compilation, or memory issues') but the bulk is technical jargon (IRs, tensors, primitives, kernels) with no file-extension or synonym enumeration, so it stops short of comprehensive.

4 / 5

Distinctiveness Conflict Risk

Highly specific to OpenVINO debug capabilities (IRs, transformations, primitives, kernels, CPU/GPU plugins), a clear niche with distinct triggers and minimal overlap risk.

5 / 5

Total

17

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 3 missing

Warning

Total

15

/

16

Passed

Repository
openvinotoolkit/openvino
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.