model-merging

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.

Quality

80%

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

68%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 highly actionable reference with executable configs throughout, weakened by redundancy between the Core Concepts and Merge Methods Guide sections and the absence of an explicit validated end-to-end workflow. Progressive disclosure is solid with clearly signaled reference files.

Suggestions

Collapse the "Core Concepts" method explanations into the "Merge Methods Guide" to remove the duplicated SLERP/Linear/Task Arithmetic configs and tighten conciseness.

Add an explicit numbered end-to-end workflow (configure YAML → run mergekit-yaml → load & benchmark → only then deploy) with a validation checkpoint before deployment, rather than burying evaluation only in Common Pitfalls.

Move the bulk of the per-method configs and the Evaluation/ProductionDeployment code into the existing reference files, keeping SKILL.md as a concise overview with one canonical example per method.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and accurate, but the "Core Concepts" section repeats all five methods that are then fully re-shown in "Merge Methods Guide", and SLERP/Linear/Task Arithmetic configs each appear multiple times — clear tightening opportunity.	3 / 5
Actionability	Provides copy-paste ready YAML configs with real model identifiers, executable commands (mergekit-yaml ... --cuda, push_to_hub), and runnable Python evaluation snippets covering the common merge cases.	5 / 5
Workflow Clarity	Sections flow logically (install → config → merge → evaluate → deploy) but there is no explicit numbered end-to-end workflow, and validation is framed as a "Common Pitfall" rather than an embedded checkpoint in the merge process.	3 / 5
Progressive Disclosure	Three real one-level-deep references (methods.md, examples.md, evaluation.md) are clearly signaled in "See Also", but the body is heavy with method/evaluation/deployment material that overlaps the reference files.	4 / 5
	Total	15 / 20 Passed

Description

92%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 strong, third-person description that clearly states what the skill does and when to use it, with concrete method names and trigger phrases. Slightly more natural-term synonym coverage would push trigger_term_quality to the top anchor.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ("Merge multiple fine-tuned models using mergekit to combine capabilities without retraining", "creating specialized models by blending domain-specific expertise") and enumerates specific methods (SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, production deployment), giving comprehensive coverage.	5 / 5
Completeness	Explicitly answers both what ("Merge multiple fine-tuned models using mergekit to combine capabilities without retraining") and when ("Use when creating specialized models... improving performance beyond single models, or experimenting rapidly with model variants") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Good natural keyword coverage ("merge models", "combine capabilities", "blend domain-specific expertise", "math + coding + chat", "experimenting rapidly with model variants") with a few natural synonyms (e.g., "fuse", "blend models") not exhaustively covered.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (model merging via mergekit with named methods like SLERP/TIES/DARE/Task Arithmetic) with distinct triggers and minimal overlap risk with unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (540 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/emerging-techniques/model-merging/SKILL.md
Commit: 4b35756

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.