unstructured-medical-text-miner

Mine unstructured clinical text from MIMIC-IV to extract diagnostic logic.

Quality

51%

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

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Evidence Insight/unstructured-medical-text-miner/SKILL.md

Quality

Content

50%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 rich and largely actionable with concrete code and schemas, but it is padded with boilerplate and duplicate sections and keeps API/spec detail inline rather than splitting it into reference files. The batch workflow also lacks an explicit output-validation feedback loop.

Suggestions

Remove placeholder cross-references ("See ## Features above") and consolidate the duplicated Workflow/Implementation Details and two References sections to cut padding.

Move the detailed Usage code, Output JSON schemas, Configuration, and CLI sections into separate reference files, leaving SKILL.md a concise overview that links out one level deep.

Add an explicit validate-output / failure-retry checkpoint to the main extraction workflow so the batch operation has a clear feedback loop.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose: placeholder pointers ("See ## Features above", "See ## Usage above"), duplicated sections (two "References", overlapping Workflow/Implementation Details), and generic risk/security/evaluation checklists that restate concepts Claude already knows.	2 / 5
Actionability	Provides mostly executable guidance — runnable Python usage, CLI invocations, a config block, and concrete input/output JSON schemas — with only minor gaps (real MIMIC data and spaCy/scispacy models required to actually execute).	4 / 5
Workflow Clarity	A sequenced workflow with a Quick Check and fallback path is present, but the batch data-mining operation lacks an explicit validate-output/failure-recovery feedback loop, capping workflow clarity at 3 per the batch-operation guideline.	3 / 5
Progressive Disclosure	There is section structure and one clearly signaled one-level-deep reference (references/audit-reference.md), but large API/IO/config/CLI content that would fit better in separate reference files is inlined in SKILL.md, so organization is only partial.	3 / 5
	Total	12 / 20 Passed

Description

53%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 states a clear, domain-specific purpose but lacks an explicit "when to use" trigger clause and broader natural keyword coverage. It is distinctive yet not comprehensive enough to score above the midpoint on most dimensions.

Suggestions

Add an explicit 'Use when...' clause naming concrete triggers (e.g., mining MIMIC-IV NOTEEVENTS, discharge summaries, or radiology reports for diagnostic reasoning).

Expand the action list to mention the concrete extraction capabilities (entities, relations, timelines, negation) so specificity reflects the body's actual features.

Include natural synonyms and formats users might say (clinical notes, NLP, .csv/.parquet) to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain (MIMIC-IV clinical text) and one concrete action ("extract diagnostic logic"), but does not enumerate multiple specific actions like entity/relation/timeline extraction that the body covers.	3 / 5
Completeness	Gives a clear "what" but has no "Use when..." or equivalent explicit trigger guidance for when to invoke it, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Contains relevant domain terms ("MIMIC-IV", "clinical text", "diagnostic logic") but misses natural variations a user might say (notes, discharge summaries, NLP, .csv) and any file extensions.	3 / 5
Distinctiveness Conflict Risk	The MIMIC-IV + diagnostic-logic niche is clearly distinct with minimal overlap risk against general document or NLP skills; it is not a 5 only because the trigger phrasing is thin.	4 / 5
	Total	13 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Evidence Insight/unstructured-medical-text-miner/SKILL.md
Commit: 63c61d3

Reviewed: 5 hours 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.