segmenting-clinical-sections

Split a clinical note into canonical sections (Chief Complaint, HPI, PMH, Medications, Allergies, Assessment & Plan, etc.) before running OpenMed NER or de-identification, so section context sharpens downstream precision. Use when the user has a free-text note or discharge summary and wants section-aware processing, header detection, mapping headers to LOINC document-section codes, or per-section NER/de-id. Covers heuristic header detection, normalization to canonical section labels, LOINC/SecTag framing, and why a finding in PMH is historical while the same finding in A&P is active. Hand-off: feed each sectioned chunk into openmed.analyze_text / openmed.deidentify. Pairs before extracting-clinical-entities.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

80%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 well-structured, actionable skill with an executable Quick start and a clear sequenced workflow. It is slightly held back by missing explicit validation/feedback loops in the workflow and minor motivational prose that could be trimmed.

Suggestions

Add an explicit validation checkpoint in the Workflow (e.g. 'Verify each chunk has a non-empty body and a recognized/canonical section label before handing off; re-check headers if chunks are empty') to close the validate->fix->retry loop.

Tighten the opening paragraph and the 'precision is won' commentary in step 4, which restate motivation already covered by the description, to improve token efficiency.

Consider pulling the SECTION_MAP and LOINC code table into a bundled reference file once the map grows, keeping SKILL.md as an overview that points one level deep.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient and assumes Claude's knowledge (e.g. it does not explain what LOINC or NER is at a basic level), but the opening paragraph and some prose in the Workflow and Hand-off sections restate motivation that could be trimmed slightly.	4 / 5
Actionability	The Quick start provides a fully executable, copy-paste-ready Python snippet with a concrete SECTION_MAP, regex, chunking loop, and a real openmed.analyze_text call covering the common case, plus specific LOINC codes throughout.	5 / 5
Workflow Clarity	The five-step Workflow is clearly sequenced with concrete details and an edge-case/gotchas section, but there is no explicit validate->fix->retry checkpoint; the per-section processing step implies but does not formalize validation of the segmentation output.	4 / 5
Progressive Disclosure	Content is well-organized into clear sections (When to use, Quick start, Workflow, Hand-off, Edge cases, Standards) with external links signaled one level deep; no bundle files exist so it is appropriately self-contained, though the Standards section is external URLs rather than local references.	4 / 5
	Total	17 / 20 Passed

Description

90%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 rich, specific description with concrete actions, explicit when-to-use triggers, and a clear clinical-niche framing that includes hand-off and pairing guidance. Minor trigger-synonym gaps (progress note, H&P as explicit triggers) keep it just short of perfect on keyword coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — splitting notes into canonical sections, header detection, normalization to canonical labels, LOINC/SecTag framing, and per-section NER/de-id hand-off — with comprehensive coverage of the skill's scope.	5 / 5
Completeness	Explicitly answers both what (split notes into canonical sections before NER/de-id, with header detection, normalization, LOINC mapping) and when ('Use when the user has a free-text note or discharge summary and wants section-aware processing...'), with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes strong natural trigger phrases ('free-text note or discharge summary', 'section-aware processing', 'header detection', 'mapping headers to LOINC'), though it omits common synonyms a user might say like 'progress note' or 'H&P' as explicit triggers.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear clinical-NLP niche (note sectioning + LOINC document-section mapping) and explicitly pairs before extracting-clinical-entities, making conflict with unrelated skills minimal and the hand-off explicit.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: maziyarpanahi/openmed
Path: skills/segmenting-clinical-sections/SKILL.md
Commit: 80da98c

Reviewed: 1 day 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.