ov-gguf-add-architecture

Enable a new model architecture/family in the OpenVINO GGUF frontend's native .gguf builder, or check whether a GGUF model is supported. Use when the user asks to enable, support or bring up a GGUF/llama.cpp model (llama, qwen, phi, gemma, MoE and similar decoder-only families), when a .gguf file is rejected as an unsupported architecture, or when working on arch_registry / DecoderBuilder in src/frontends/gguf/src/builder/. Do NOT use for adding a single ggml op translator or for debugging wrong output from an already-supported model.

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

86%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 compact, well-structured instruction skill that defers detail to one-level-deep docs while keeping the inline workflow concrete and verifiable. Its only weakness is that the deepest executable detail rests in the referenced documents rather than the body itself.

Dimension	Reasoning	Score
Conciseness	A lean four-step body with no padding and no re-explanation of concepts Claude already knows; every line carries actionable signal about files, symbols, and checks.	5 / 5
Actionability	Concrete guidance names exact files (adding_an_architecture.md, supported_models.md), symbols (arch_registry.cpp, RoPE type), and checks (graph-fingerprint check), but the fully executable steps are deferred to the referenced docs.	4 / 5
Workflow Clarity	A clear numbered sequence with an explicit verification checkpoint ('Verify as that document prescribes, including the graph-fingerprint check') and a re-run feedback loop after shared-logic changes, though detailed recovery steps live in the external doc.	4 / 5
Progressive Disclosure	A short single-purpose overview with clearly signaled one-level-deep references to dedicated docs; detail is appropriately split out rather than inlined, and navigation is easy.	5 / 5
	Total	18 / 20 Passed

Description

87%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 tightly scoped, well-triggered description that clearly states capabilities, concrete use-when conditions, and out-of-scope boundaries. It is among the stronger examples, with only minor keyword-coverage gaps.

Dimension	Reasoning	Score
Specificity	Names two concrete actions ('Enable a new model architecture/family', 'check whether a GGUF model is supported') plus specific internal symbols (arch_registry, DecoderBuilder), but coverage is narrow to one builder area, leaving minor gaps.	4 / 5
Completeness	Explicitly answers both what (enable/check-support) and when via a concrete 'Use when...' clause plus negative 'Do NOT use for...' scope, matching the clearest anchor.	5 / 5
Trigger Term Quality	Good natural-term coverage including 'GGUF/llama.cpp model', family names (llama, qwen, phi, gemma, MoE), and the '.gguf file rejected as an unsupported architecture' phrasing, though a few common synonyms are absent.	4 / 5
Distinctiveness Conflict Risk	Highly specific niche (OpenVINO GGUF frontend builder, arch_registry/DecoderBuilder) with explicit negative-scope guidance, yielding minimal overlap with adjacent skills.	5 / 5
	Total	18 / 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 suspicious	Warning

	Total	15 / 16 Passed

Repository: openvinotoolkit/openvino
Path: .claude/skills/ov-gguf-add-architecture/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.