minicpm5-deploy-vllm

Serve MiniCPM5-1B via vLLM as an OpenAI-compatible HTTP server. Use when the user wants high-throughput production serving on NVIDIA GPU, asks for "vLLM", "OpenAI server", "REST API for MiniCPM5", or "production deployment".

Quality

85%

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

88%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, well-structured deployment skill with executable commands, a clear validated workflow, and tight prose. The only notable weaknesses are mild verbosity in the tool-calling version narrative and the absence of any bundled detailed reference to lean on.

Suggestions

Trim the dated PR/version narrative in the tool-calling section to the essential operational fact (use the bundled plugin until a release ships the parser natively) to improve conciseness.

Add a bundled references/ or scripts/ file for the tool-parser plugin and the sampling/pitfall details, and link to it from SKILL.md so progressive disclosure relies on verified bundle paths rather than an external repo doc.

Verify the docs/deployment/vllm.md link resolves within the skill bundle, or convert it to a bundled reference file.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence with no concept explanations, but the tool-calling paragraph carries a somewhat verbose dated PR/version narrative ("merged to main on 2026-05-27", "v0.22.0 was cut before the merge") that could be trimmed, keeping it just below the lean-and-efficient anchor.	4 / 5
Actionability	Provides fully executable, copy-paste-ready commands for install, launch, and validation with concrete env-var substitution and an expected-output curl example covering the common case.	5 / 5
Workflow Clarity	Steps are clearly sequenced (Install -> Launch -> Validate) with an explicit validation gate ("Wait for Application startup complete") and an error-recovery hint (the <think> tag check), giving explicit checkpoints and a feedback loop.	5 / 5
Progressive Disclosure	Sections are well-organized with a clearly signaled one-level-deep reference to docs/deployment/vllm.md, but no bundled reference/script/asset files exist and the single link points outside the skill bundle, leaving a minor organization gap.	4 / 5
	Total	18 / 20 Passed

Description

82%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, concise description that clearly states a specific capability and pairs it with explicit, natural trigger phrases and a well-scoped niche. The only minor gap is that it describes a single primary action rather than enumerating several distinct capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain and a specific concrete action ("Serve MiniCPM5-1B via vLLM as an OpenAI-compatible HTTP server"), but essentially describes one serving action rather than a list of several distinct actions, so it sits at the 1-2-concrete-actions anchor rather than the several-actions anchor.	3 / 5
Completeness	It explicitly answers both what (serve MiniCPM5-1B via vLLM as an OpenAI-compatible HTTP server) and when ("Use when the user wants high-throughput production serving on NVIDIA GPU, asks for...") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural user phrases like "vLLM", "OpenAI server", "REST API for MiniCPM5", and "production deployment", giving good keyword coverage, though a few natural synonyms (e.g. "API server", file extensions) are absent.	4 / 5
Distinctiveness Conflict Risk	The niche is narrow and specific (MiniCPM5-1B + vLLM + NVIDIA production serving) with distinct triggers, making conflict with other skills minimal.	5 / 5
	Total	17 / 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-vllm/SKILL.md
Commit: 719e4fc

Reviewed: 17 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.