ov-debug-matcher-pass

Debug why an OpenVINO MatcherPass transformation is not firing. Use this skill immediately when a user says a transformation is "not applied", a "pass has no effect", a "matcher never triggers", a pattern "doesn't match", a "callback never fires", "WrapType predicate is too strict", a subgraph "not fused" despite the pass being registered, or they see "END: PATTERN DIDN'T MATCH" in matcher logs. Also trigger when a MatcherPass works on one model but silently skips another, when the user wants to add a reproducer test for a transformation that should fire but doesn't, or when they suspect an opset version mismatch preventing a match. Do NOT trigger for: writing a new MatcherPass from scratch, debugging a pass that fires but produces wrong numerical results, crashes in pass registration, or general questions about what MatcherPass is.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

77%

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

The body is highly actionable with a clearly sequenced, validated workflow, but it carries some duplicated root-cause content and keeps a large amount of detail inline rather than splitting it into reference files. Tightening the overlap and moving some detail to references would lift the conciseness and progressive-disclosure scores.

Suggestions

Consolidate the 3c log-phrase table and the 3d root-cause list into a single mapping to remove duplicated reasoning and tighten conciseness.

Move the full Step 5 C++ test template and/or the Step 6 resolution table into a reference file (e.g., references/reproducer-template.md) so SKILL.md stays an overview, improving progressive disclosure.

Move the large 'Common root causes' enumeration behind a clearly signaled reference while keeping a short decision summary inline, reducing body length without losing actionability.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and task-specific, but the root-cause material is duplicated across the 3c table and the 3d numbered list with overlapping content, and several explanatory notes could be consolidated; not quite the lean 'every token earns its place' bar.	2 / 3
Actionability	Provides fully executable guidance: concrete grep/cmake/env-var commands, a runnable Python tally snippet, and a copy-paste C++ test template with its semantics explained — copy-paste ready.	3 / 3
Workflow Clarity	Clear sequenced Steps 0–6 with an early-exit condition and explicit feedback loops — the reproducer is cross-validated against matcher logs with 'if logs diverge, revise and repeat', a strong validation checkpoint.	3 / 3
Progressive Disclosure	The one reference file (example-diagnosis-report.md) is clearly signaled and one level deep, but the bulk of detail (root-cause and resolution tables, full test template) lives inline in a fairly monolithic body rather than being split into separate reference files.	2 / 3
	Total	10 / 12 Passed

Description

100%

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, trigger-rich, and complete, clearly stating both what the skill does and exactly when to invoke it, with negative triggers that sharply bound its scope. It is an excellent, low-conflict trigger description.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — diagnosing why a MatcherPass does not fire and producing a reproducer test — matching the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers both what (debug non-firing MatcherPass + add reproducer test) and when via 'Use this skill immediately when...' plus 'Also trigger when...' and negative 'Do NOT trigger for...' guidance.	3 / 3
Trigger Term Quality	Covers natural phrasings a user would actually say ('not applied', 'pass has no effect', 'matcher never triggers', 'doesn't match', 'callback never fires', 'not fused'), giving good coverage of natural terms.	3 / 3
Distinctiveness Conflict Risk	Targets a narrow niche (OpenVINO MatcherPass non-firing) with explicit negative triggers that rule out adjacent tasks, making conflict with other skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 suspicious	Warning

	Total	15 / 16 Passed

Repository: openvinotoolkit/openvino
Path: .claude/skills/ov-debug-matcher-pass/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.