minicpm5-deploy-mlx

Run MiniCPM5-1B or MiniCPM5-2B natively on Apple Silicon with Apple's MLX framework. Use when the user has an Apple Silicon Mac and asks for "MLX", "mlx_lm", "mlx_lm.convert", "mlx_lm.generate", or wants the fastest path on Apple Silicon.

Quality

91%

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

92%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 is concise, highly actionable, and workflow-clear with explicit validation and recovery guidance. The main weakness is a broken external reference link that resolves to a non-existent file.

Suggestions

Fix or remove the Reference link to ../../docs/deployment/mlx.md, which does not exist; either create the referenced file or drop the link.

Consider adding the model names ('MiniCPM5-1B', 'MiniCPM5-2B') as explicit trigger terms in the description to round out trigger coverage.

If a deployment details doc is intended, create it and verify the relative path resolves from the skill directory.

Dimension	Reasoning	Score
Conciseness	Lean and command-focused with no padding of concepts Claude already knows; even the one-line intro earns its place by contrasting against llama.cpp/vLLM rather than explaining MLX.	5 / 5
Actionability	Fully executable, copy-paste-ready commands cover the common cases: install, pre-converted generation, local bf16/4-bit conversion, and an OpenAI-compatible server with a complete curl example.	5 / 5
Workflow Clarity	A clear numbered sequence (1 install, 2A/2B run, 3 validate) with an explicit validation checkpoint ('The reply should contain "2"') plus Common pitfalls that provide error-recovery feedback (eos token override, warm-cache note).	5 / 5
Progressive Disclosure	Section structure is clean and well-organized, but the single external reference (../../docs/deployment/mlx.md) points to a file that does not exist and no bundle files are present, so the reference target is broken rather than merely a minor gap.	3 / 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.

The description is specific, well-triggered, and distinct, cleanly answering both what the skill does and when to invoke it. Third-person voice is maintained throughout.

Dimension	Reasoning	Score
Specificity	Names the specific models, hardware, and framework ('Run MiniCPM5-1B or MiniCPM5-2B natively on Apple Silicon with Apple's MLX framework') plus tool-level action cues, but does not enumerate distinct actions like convert/serve in the description itself, leaving minor gaps below comprehensive coverage.	4 / 5
Completeness	Explicitly answers both what ('Run MiniCPM5-1B or MiniCPM5-2B natively on Apple Silicon with Apple's MLX framework') and when ('Use when the user has an Apple Silicon Mac and asks for...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural-term coverage including exact tool names users say ('MLX', 'mlx_lm', 'mlx_lm.convert', 'mlx_lm.generate') and a hardware condition, but misses synonyms such as 'MiniCPM5' as a standalone trigger or '.mlx' file extensions.	4 / 5
Distinctiveness Conflict Risk	A clear narrow niche tied to specific models, a specific framework, and Apple Silicon hardware, with sibling deployment skills distinguished elsewhere, giving minimal conflict risk.	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: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: OpenBMB/MiniCPM
Path: skills/minicpm5-deploy-mlx/SKILL.md
Commit: 316cfb1

Reviewed: 15 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.