CtrlK
BlogDocsLog inGet started
Tessl Logo

loading-openmed-models

Load OpenMed clinical/biomedical NER models from the Hugging Face Hub or a local path and reuse them efficiently across calls. Use when the user wants to load an OpenMed model, control the model cache, run fully offline after a one-time download, reuse a ModelLoader to avoid reloading, set a cache_dir or device, or pick between a registry key, a full Hugging Face id, and a local directory. Pairs with choosing-openmed-models (pick the model) and extracting-clinical-entities (run it).

77

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

93%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 lean, executable, and well-structured for a focused model-loading skill. Code is copy-paste ready and edge cases are surfaced as actionable gotchas. The only gap is the absence of explicit validation checkpoints, which matters little here since loading models is non-destructive.

Suggestions

Add a brief validation checkpoint after the offline/cache setup (e.g. confirm `loader.loaded_models()` shows the expected key before proceeding in a batch pipeline) to push workflow_clarity toward 5.

Trim the opening paragraph, which restates scope already covered by the description, to tighten conciseness further.

Consider a one-line 'verify offline works' check after the HF_HUB_OFFLINE/TRANSFORMERS_OFFLINE export block so the offline workflow has an explicit success signal.

DimensionReasoningScore

Conciseness

Lean throughout: tight code examples, a compact naming-form table, and an intro that assumes Claude knows what caches/tokenizers/NER are — no padding about library basics. The only minor redundancy is the intro echoing the description's scope, but it is brief.

5 / 5

Actionability

Fully executable, copy-paste-ready code covers the common cases: install, ModelLoader reuse in a loop, load_model bundles, OpenMedConfig, offline env vars, local-path loading, get_model_max_length, and unload — each a complete runnable snippet.

5 / 5

Workflow Clarity

Each task is clearly sequenced (e.g. '1. First run online; 2. Every run after offline') and the Edge cases section provides error-recovery guidance (force_reload, path/key collision, private repos). It stops short of explicit validation checkpoints/feedback loops, which the anchor 5 requires — though model loading is non-destructive so the destructive-cap does not apply.

4 / 5

Progressive Disclosure

Well-organized into focused sections with one-level-deep, clearly signaled references — sibling skills (choosing-openmed-models, extracting-clinical-entities) and external HF doc URLs. No nested references and no content inlined that clearly belongs in a separate bundle file (none exist).

5 / 5

Total

19

/

20

Passed

Description

100%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 third-person, specific, and answers both what and when with concrete trigger phrases. It cleanly demarcates its scope from adjacent OpenMed skills. No vague fluff or over-claims.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'Load OpenMed clinical/biomedical NER models from the Hugging Face Hub or a local path', 'reuse them efficiently across calls', 'control the model cache', 'run fully offline after a one-time download', 'set a cache_dir or device', 'pick between a registry key, a full Hugging Face id, and a local directory' — with comprehensive coverage of the skill's capabilities.

5 / 5

Completeness

Explicitly answers both 'what' (load/reuse OpenMed NER models from Hub or local path) and 'when' via a concrete 'Use when the user wants to...' clause enumerating specific trigger conditions.

5 / 5

Trigger Term Quality

Natural trigger phrases a user would actually say — 'load an OpenMed model', 'control the model cache', 'run fully offline', 'reuse a ModelLoader to avoid reloading', 'set a cache_dir or device' — plus synonym coverage across registry key / HF id / local directory naming forms.

5 / 5

Distinctiveness Conflict Risk

Clear niche (loading OpenMed NER models) with explicit disambiguation from sibling skills — 'Pairs with choosing-openmed-models (pick the model) and extracting-clinical-entities (run it)' — minimizing conflict risk.

5 / 5

Total

20

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
maziyarpanahi/openmed
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.