add-fe-op

Adds a new operation to OpenVINO Frontend pipelines with translator updates, registration, and tests.

Quality

69%

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

Fix and improve this skill with Tessl

tessl review fix ./.github/agents-prototype/skills/add-fe-op/SKILL.md

Quality

Content

63%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 presents a lean, well-sequenced instruction workflow with genuinely useful OpenVINO-specific translation guidance and explicit validation gates. Its two real weaknesses are dangling references (pytorch.md, onnx.md, conversion-issues files are absent from the bundle) and steps that describe what to do without the concrete commands or file paths needed to execute them.

Suggestions

Ship the referenced files (pytorch.md, onnx.md, conversion-issues/onnx.md, conversion-issues/pytorch.md) in the bundle or inline their key steps — every reference in the body is currently a dead link.

Add concrete commands/file paths for the validation steps, e.g. the actual build targets for frontend changes and the conversion-check command for each frontend, and fix the lowercase sub-bullets in step 6.

Merge the two '## Notes' sections into one to remove the duplicated 'Keep this skill instruction-only in markdown' line.

Dimension	Reasoning	Score
Conciseness	The body is instruction-focused with no explanations of concepts Claude already knows, matching 'efficient; minor instances that could be trimmed'. Not 5 because of duplication: a '## Notes' section with 'Keep this skill instruction-only in markdown' appears twice (before and after the workflow), and the second Notes repeats 'Prefer minimal, root-cause updates'.	4 / 5
Actionability	Concrete anchors exist ('ONNX_OP macro registration', 'ShapeOf-based graph logic', 'aten::op' / 'aten.op.default' examples), but many steps remain descriptive rather than executable — e.g. 'run conversion check from generated <frontend> model' gives no command, and steps like 'Add converter declaration in op table' name no file paths. This fits 'some concrete guidance but incomplete; missing key details'; not 4 because there is no executable code or command anywhere in the body.	3 / 5
Workflow Clarity	A clear numbered 6-step sequence with explicit checkpoints ('If neither is true, do not report conversion status as passed; mark validation as blocked', 'Confirm rerun is idempotent', three-state final reporting). Fits 'clear sequence with most checkpoints present; minor validation gaps'; not 5 because there is no fix-and-retry error-recovery loop when validation fails, and step 6 begins with lowercase, inconsistently formatted sub-bullets.	4 / 5
Progressive Disclosure	References are clearly signaled via tables ('pytorch.md', 'onnx.md', '../conversion-issues/*.md') with what each covers, but none of these files exist in the bundle — there is no references/ directory and the linked paths are dangling, so navigation actually fails. The TensorFlow/generic workflow is also inlined in the main file. This fits 'references present but not clearly signaled / content that should be separate is inline' territory better than the 4 anchor, whose references resolve.	3 / 5
	Total	14 / 20 Passed

Description

62%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 terse, third-person, domain-specific description that clearly states what the skill does, but it lacks any trigger guidance ('Use when...') and natural-language variations a user might say. Distinctiveness is excellent thanks to the OpenVINO niche.

Suggestions

Add an explicit trigger clause, e.g. 'Use when adding or enabling a new op in OpenVINO frontend support (PyTorch, ONNX, TensorFlow).' This lifts completeness from 3 to 4-5.

Include natural user phrasings and synonyms such as 'add op support', 'new converter', 'frontend op registration' to improve trigger_term_quality.

Mention conversion validation in the action list, since the body treats it as a required final step.

Dimension	Reasoning	Score
Specificity	Names the domain ("OpenVINO Frontend pipelines") and several concrete actions ("translator updates, registration, and tests"), matching the 'several specific actions; minor gaps' anchor. Not 5 because it omits coverage of conversion validation and the supported frontends (PyTorch/ONNX) that the body actually handles.	4 / 5
Completeness	The 'what' is clear ("Adds a new operation ... with translator updates, registration, and tests") but there is no 'when' / 'Use when' clause at all, so the explicit cap of 3 applies. Not 2 because the 'what' half is concrete and unambiguous.	3 / 5
Trigger Term Quality	Relevant keywords like "OpenVINO", "operation", "translator", "registration" are present and a user would plausibly say "add a new operation to OpenVINO", but common variations are missing ("op", "converter", "PyTorch", "ONNX", "frontend op support"). Fits 'some relevant keywords but missing common variations' rather than the 4 anchor's good coverage.	3 / 5
Distinctiveness Conflict Risk	"OpenVINO Frontend" is a clear niche with distinct triggers; virtually no other skill would claim these terms, so conflict risk is minimal. Not 4 because there is no meaningful overlap with adjacent skills in the description itself.	5 / 5
	Total	15 / 20 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: 2 missing, 2 suspicious	Warning

	Total	15 / 16 Passed

Repository: openvinotoolkit/openvino
Path: .github/agents-prototype/skills/add-fe-op/SKILL.md
Commit: f766feb

Reviewed: 3 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.