CtrlK
BlogDocsLog inGet started
Tessl Logo

conversion-issues

Investigate and fix model conversion issues in OpenVINO Frontends (ONNX, PyTorch) — triage, debugging, accuracy comparison, and pre-submission verification.

60

Quality

70%

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 ./.github/agents-prototype/skills/conversion-issues/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 an exemplary lean routing document with clear navigation and explicit validation checkpoints, assuming Claude's intelligence throughout. Its weakness is that the core actionable debugging content lives in referenced files not present in the evaluated bundle.

Suggestions

Add a brief inline quick-start (e.g., one triage decision tree or the reference-runtime verification command) so the body is actionable even before opening the referenced files.

Ensure onnx.md and pytorch.md ship in the bundle (or move them under references/) so the signaled one-level-deep references resolve.

Surface the fix→test→retry feedback loop explicitly in the body rather than only in Notes, to make the validation sequence a first-class checklist.

DimensionReasoningScore

Conciseness

Lean and efficient: a ~30-line routing document that assumes Claude's competence, never explains what OpenVINO/PDFs/frontends are, and has no padding — every token earns its place.

5 / 5

Actionability

The body gives concrete routing (pick the matching frontend file) and three concrete rules in Notes, but the substantive debugging steps are deferred to referenced files that are not present in the bundle, leaving key execution details missing.

3 / 5

Workflow Clarity

A clear high-level sequence (route to frontend workflow → verify reference runtime first → root-cause fix → add test → pass full suite) with explicit validation checkpoints in Notes; the detailed fix→test→retry loop lives in the referenced files, a minor gap.

4 / 5

Progressive Disclosure

Well-structured overview with tables that signal one-level-deep references and describe what each covers, but the core referenced files (onnx.md, pytorch.md) are absent from the bundle, so the split cannot be verified against actual bundle structure.

4 / 5

Total

16

/

20

Passed

Description

66%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 specific and distinct, naming a clear niche and several concrete actions, with good natural trigger terms. Its main weakness is the missing explicit "Use when..." trigger guidance, which caps completeness.

Suggestions

Add an explicit "Use when..." clause naming concrete triggers (e.g., "Use when a model fails to convert to OpenVINO IR via the ONNX or PyTorch frontend, or produces wrong inference results").

Include file extensions and synonyms (".onnx", ".pt", "IR", "compile/convert") to broaden natural trigger coverage.

Tighten the abstract "fix" with a concrete action (e.g., "patch the op translator") to push specificity toward comprehensive coverage.

DimensionReasoningScore

Specificity

Lists several concrete actions — "Investigate and fix", "triage, debugging, accuracy comparison, and pre-submission verification" — covering the workflow, though "fix" stays somewhat abstract, leaving minor gaps versus a fully comprehensive action list.

4 / 5

Completeness

The "what" is clear and specific, but there is no "Use when..." clause or equivalent explicit trigger guidance, which per the rubric caps completeness at 3; "when" is only weakly implied.

3 / 5

Trigger Term Quality

Strong domain keywords a user would naturally say ("model conversion issues", "OpenVINO", "ONNX", "PyTorch", "triage", "debugging"), but missing file extensions (.onnx, .pt) and synonyms like "compile" or "IR".

4 / 5

Distinctiveness Conflict Risk

A clear niche (OpenVINO frontend conversion issues for ONNX/PyTorch) with distinct triggers, but minor overlap risk with the related add-fe-op skills and no explicit trigger clause to fully separate them.

4 / 5

Total

15

/

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: 2 missing, 2 suspicious

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.