CtrlK
BlogDocsLog inGet started
Tessl Logo

minicpm5-deploy-transformers

Run MiniCPM5-1B with Hugging Face Transformers for one-shot Python generation on GPU (bfloat16) or CPU (float32). Use when the user wants a quick Python script, no server, no extra deps, or asks for "Transformers", "AutoModelForCausalLM", "model.generate" with MiniCPM5.

69

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

82%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.

An efficient, highly actionable deploy skill with executable code, a validation step, and clean organization. Main gaps are inline version-pin time-sensitivity and a missing validation failure-recovery loop.

Suggestions

Move version pins into a dedicated 'Compatibility / fallback versions' subsection or a referenced file so the main install line stays free of time-sensitive specifics.

Add a brief failure-recovery note under Validate (e.g., what to check if output is incoherent — dtype/device, chat template, token limits) to close the workflow feedback loop.

Consider extracting the LoRA inference block into a short referenced file or clearly flagging it as optional so the core run path reads even leaner.

DimensionReasoningScore

Conciseness

Lean body with no concept over-explanation and tight tables, but inline time-sensitive version pins ('transformers>=5.6,<6', 'torch>=2.11') are not isolated in a deprecated/old-patterns section, which the guideline penalizes below the 5 anchor.

4 / 5

Actionability

Copy-paste-ready install commands and a complete executable Python script, plus a CPU variant, a LoRA snippet, and a sampling-defaults table cover the common cases, matching the 5 anchor.

5 / 5

Workflow Clarity

Clear two-step sequence (Install, Run) with an explicit Validate checkpoint ('1+1=?' expected output), but no error-recovery feedback loop is specified, keeping it below the 5 anchor.

4 / 5

Progressive Disclosure

Well-organized sections with a single clearly signaled one-level reference (docs/deployment/transformers.md); no bundle files exist to split into, so it is well-structured but does not fully realize the multi-file navigation of the 5 anchor.

4 / 5

Total

17

/

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 tight, well-targeted description that clearly states capability and usage triggers with concrete API keywords. Minor room only in breadth of action coverage and synonym richness.

DimensionReasoningScore

Specificity

Names the domain and concrete setup ('MiniCPM5-1B with Hugging Face Transformers for one-shot Python generation on GPU (bfloat16) or CPU (float32)') but covers essentially one action with two environment variants rather than multiple distinct actions, so it sits below the comprehensive 5 anchor.

4 / 5

Completeness

Explicitly answers both what (run MiniCPM5-1B via HF Transformers, one-shot generation, GPU/CPU) and when ('Use when the user wants a quick Python script...') with concrete trigger phrases, matching the 5 anchor.

5 / 5

Trigger Term Quality

Strong keyword coverage combining natural phrasing ('quick Python script, no server, no extra deps') with API symbols users cite ('Transformers', 'AutoModelForCausalLM', 'model.generate'), though it lacks synonym/extension breadth that would mark the 5 anchor.

4 / 5

Distinctiveness Conflict Risk

Carves a clear niche (MiniCPM5 + Transformers, no server, no extra deps) distinct from sibling vllm/sglang/mlx/llama-cpp deploy skills, minimizing wrong-skill triggering.

5 / 5

Total

18

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 suspicious

Warning

Total

15

/

16

Passed

Repository
OpenBMB/MiniCPM
Reviewed

Table of Contents

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.