verify-conversion

E2E gate — verifies that applied patches produce a working, numerically sane end-to-end inference through the OpenVINO plugin. Handles HuggingFace/optimum-intel, native OV conversion (ovc/convert_model), and ONNX. Used by orchestrators as the mandatory gate before any PR is published.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

92%

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

A highly actionable, concise gate skill with explicit validation checkpoints and concrete code across all conversion paths. Its main weakness is that all detail lives inline with no progressive disclosure into separate reference files.

Suggestions

Fix the step numbering gap (Step 1 jumps to Step 3) so the workflow sequence is unambiguous.

Consider moving the large PIPELINE_TAG_MAP and per-path code examples into a references/ file, keeping SKILL.md as a lean overview with one-level-deep pointers.

Add a brief explicit retry/fallback note for Path B failures to mirror the Path A retry guidance.

Dimension	Reasoning	Score
Conciseness	The body is lean — decision tables and executable code blocks dominate, with no padding explaining what OpenVINO/ONNX are — so nearly every token earns its place.	3 / 3
Actionability	Provides copy-paste-ready Python and bash, a signal-to-path decision table, and exact output paths (e.g. agent-results/<agent>/verify_result.json); the lone torch_model placeholder is explicitly justified.	3 / 3
Workflow Clarity	Sequences Determine-path -> convert -> sanity check -> report with an explicit validation checkpoint and a feedback directive ('If the sanity check fails... do not silently swallow the failure'), matching the top anchor despite the Step 1->Step 3 numbering gap.	3 / 3
Progressive Disclosure	No bundle files exist, so everything is inline in one ~217-line SKILL.md; sections are well-organized but content like the full PIPELINE_TAG_MAP and per-path code could be split into reference files, leaving structure good but not maximally disclosed.	2 / 3
	Total	11 / 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 specific, well-scoped description that conveys both the concrete actions and the mandatory-gate trigger context. It is slightly jargon-forward in its trigger terms but otherwise clearly distinguishes the skill's purpose.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'verifies that applied patches produce a working, numerically sane end-to-end inference through the OpenVINO plugin' — and names specific tools ('optimum-intel', 'ovc/convert_model', 'ONNX'), matching the top anchor.	3 / 3
Completeness	Clearly answers 'what' and provides equivalent explicit trigger guidance via 'Used by orchestrators as the mandatory gate before any PR is published', satisfying both halves rather than leaving 'when' merely implied.	3 / 3
Trigger Term Quality	Contains relevant domain terms ('OpenVINO', 'optimum-intel', 'ONNX', 'inference', 'PR') but they skew technical and miss common natural variations a user might say, so coverage is partial rather than comprehensive.	2 / 3
Distinctiveness Conflict Risk	Occupies a distinct niche (OpenVINO conversion E2E verification gate with named conversion paths) with triggers unlikely to overlap with unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: openvinotoolkit/openvino
Path: .github/agents-prototype/skills/verify-conversion/SKILL.md
Commit: 4c20980

Reviewed: 4 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.