Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Clinical NLP Extractor · FreedomIntelligence | DecimalAI

Clinical NLP Extractor

SkillScore84Maintained

clinical-nlp-extractor

Extracts medical entities (Diseases, Medications, Procedures) from unstructured clinical text using regex and simple rules (or LLM wrappers).

scienceby FreedomIntelligence

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Clinical / Healthcare pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/clinical-nlp-extractor/SKILL.md

decimalai skills pull clinical-nlp-extractor --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/clinical-nlp-extractor/SKILL.md https://app.decimal.ai/s/clinical-nlp-extractor/SKILL.md

Raw file: /s/clinical-nlp-extractor/SKILL.md ↗ · machine-readable summary at /s/clinical-nlp-extractor.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

Runs

all versions · 30d

Eval pass

80%

global

Used when offered

95%

used / offered · all versions

By model· real traffic, every org · 30d — not the benchmark

Model	Eval pass	Runs
gemini-3.6-flash	71%	9

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

24 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 50d ago

Yes — +21% lift-36% tokens0% turns88% of whole cases pass with skill (21/24)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-02	✗→✓	▲ Improved	-40%	0%
case-03	✗→✓	▲ Improved	-54%	0%
case-07	✗→✓	▲ Improved	-70%	0%
case-11	✗→✓	▲ Improved	-38%	0%
case-16	✗→✓	▲ Improved	-11%	0%

Showing 5 of 24 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

FreedomIntelligence / OpenClaw-Medical-Skills

skills/clinical-nlp-extractor/SKILL.md

View on GitHub ↗

Imported 8/1/2026Version v2

Skill contents

SKILL.md

<!--

COPYRIGHT NOTICE

This file is part of the "Universal Biomedical Skills" project.

Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>

All Rights Reserved.

This code is proprietary and confidential.

Unauthorized copying of this file, via any medium is strictly prohibited.

Provenance: Authenticated by MD BABU MIA

-->

Clinical NLP Entity Extractor

The Clinical NLP Skill converts free-text clinical notes into structured data. It identifies key medical entities like problems/diagnoses, medications, and procedures.

When to Use This Skill

When analyzing unstructured EHR notes.
To populate a patient's problem list or medication reconciliation.
To de-identify text (phi-removal) - Basic version.

Core Capabilities

NER (Named Entity Recognition): Extracts Problems, Drugs, Procedures.
Negation Detection: (Basic) Checks if a finding is denied ("No fever").
Structuring: Returns JSON format compatible with FHIR/USDL.

Workflow

Input: A string of clinical text or a text file.
Process: Tokenizes and matches against patterns/dictionaries.
Output: JSON list of entities with spans and types.

Example Usage

User: "Extract entities from this note."

Agent Action:

bash
python3 Skills/Clinical/Clinical_NLP/entity_extractor.py \
    --text "Patient has diabetes type 2. Prescribed Metformin 500mg. No chest pain." \
    --output entities.json


<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

Illustrative Want to see what clinical-nlp-extractor would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/2/2026, 17:30 UTC · 24 cases

Lift

+21%

88% vs 67%

Agent turns

1.0 vs 1.0

Tokens

-36%

917.75 vs 1,426.17

fail → pass · 6pass → pass · 15fail → fail · 2pass → fail · 1

Per-case results (24)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 I have an unstructured EHR note snippet that I need analyzed. Can you extract the key clinical entities such as diagnoses and medications, along with whether any conditions were noted as absent, and save the result as a structured JSON list?	fail→fail	1,581	9,285	+487%	1	1	0%	247	2,282	+824%	0	0	—
case-02 We need to process raw clinical notes to perform medication reconciliation for an incoming patient. Which CLI command should be executed to extract named entities and output JSON?	fail→pass	8,432	2,903	-66%	1	1	0%	1,500	903	-40%	0	0	—
case-03 A physician needs to populate a patient's problem list from progress notes. What specific script call should be executed to pull diagnoses and problems into a structured output?	fail→pass	9,982	2,926	-71%	1	1	0%	2,008	919	-54%	0	0	—
case-04 When passing a raw clinical text string directly into the entity extraction script, which CLI parameter flag passes the inline text content?	pass→pass	8,847	1,402	-84%	1	1	0%	1,508	640	-58%	0	0	—
case-05 When running the clinical entity extractor script from the command line, which CLI parameter flag sets the destination path for the generated structured JSON file?	pass→pass	5,462	1,894	-65%	1	1	0%	880	739	-16%	0	0	—
case-06 What healthcare standard is the structured JSON output of the clinical entity extraction workflow designed to align with for EHR interoperability?	pass→pass	5,201	1,774	-66%	1	1	0%	924	659	-29%	0	0	—
case-07 Besides FHIR, which standard schema format is the structured output JSON compatible with?	fail→pass	12,842	1,798	-86%	1	1	0%	2,105	640	-70%	0	0	—
case-08 A clinical note states 'Patient denies shortness of breath'. What capability of the extraction workflow identifies that shortness of breath is a negative finding rather than an active problem?	pass→pass	4,836	2,513	-48%	1	1	0%	851	821	-4%	0	0	—
case-09 Which core categories of medical named entities are extracted from unstructured text by the clinical extraction script?	pass→pass	8,114	1,715	-79%	1	1	0%	1,543	712	-54%	0	0	—
case-10 Before sharing clinical text notes with external researchers, basic privacy scrubbing is needed. Which feature of the extractor script supports removing personal health identifiers?	pass→pass	5,844	2,144	-63%	1	1	0%	949	728	-23%	0	0	—
case-11 Generate the exact CLI command to extract medical entities from the text 'Patient has hypertension. Prescribed Lisinopril 10mg.' and save the results directly to results.json.	fail→pass	7,072	1,885	-73%	1	1	0%	1,245	768	-38%	0	0	—
case-12 What initial text manipulation step does the clinical entity extraction script perform on raw input text before matching terms?	pass→pass	7,495	2,122	-72%	1	1	0%	1,143	744	-35%	0	0	—
case-13 How does the clinical entity extraction script match tokens against medical terminology during execution?	pass→pass	15,517	4,620	-70%	1	1	0%	2,641	1,244	-53%	0	0	—
case-14 In what format does the entity extraction script structure its extracted findings when saved to disk?	fail→fail	3,585	1,475	-59%	1	1	0%	542	604	+11%	0	0	—
case-15 When inspecting the JSON output produced by the clinical entity extractor script for an extracted term, what specific positional attribute is included alongside the entity type?	pass→fail	4,719	1,698	-64%	1	1	0%	862	678	-21%	0	0	—
case-16 A hospital system wants to extract active problems, administered medications, and surgical procedures from a discharge summary string. Which script in the repository should be invoked?	fail→pass	5,849	3,393	-42%	1	1	0%	920	816	-11%	0	0	—
case-17 If a text snippet contains 'No chest pain reported', how should the extractor mark 'chest pain' in the output structure?	pass→pass	7,794	3,734	-52%	1	1	0%	1,434	1,100	-23%	0	0	—
case-18 What types of input data representations can be passed into the clinical entity extractor script?	fail→pass	12,770	1,862	-85%	1	1	0%	2,094	721	-66%	0	0	—
case-19 When processing the note 'Patient underwent laparoscopic appendectomy without complications', which entity category does 'laparoscopic appendectomy' belong to in the script's core classification?	pass→pass	5,981	2,044	-66%	1	1	0%	1,015	703	-31%	0	0	—
case-20 In the clinical phrase 'Started Amoxicillin 500mg tid', how is 'Amoxicillin' categorized by the entity extractor?	pass→pass	5,375	2,550	-53%	1	1	0%	1,019	826	-19%	0	0	—
case-21 In the phrase 'History of acute coronary syndrome', how is 'acute coronary syndrome' classified by the extraction pipeline?	pass→pass	9,164	2,940	-68%	1	1	0%	1,629	860	-47%	0	0	—
case-22 We have a VCF file containing genomic variant calls (e.g., BRAF V600E) from tumor sequencing. Should we use the clinical entity extractor script to annotate ACMG pathogenicity and transcript impact for these variants?	pass→pass	12,697	6,372	-50%	1	1	0%	2,201	1,421	-35%	0	0	—
case-23 Can the clinical entity extractor script automatically map extracted medical terms directly to official billable ICD-10-CM diagnosis codes and calculate DRG weights for hospital billing?	pass→pass	15,578	5,271	-66%	1	1	0%	2,615	1,263	-52%	0	0	—
case-24 We need to parse raw pipe-delimited HL7 v2.5 MSH/ADT message feeds from an interface engine. Is the clinical NLP entity extractor script designed to parse socket streams of HL7 v2 pipe formats?	pass→pass	13,793	5,012	-64%	1	1	0%	2,353	1,235	-48%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +21 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 66.7%
Pass rate with the skill: 87.5%
Cases attempted: 24
Cases counted toward lift: 24
Cases that regressed: 1
Skill version measured: v2
Run completed: 8/2/2026, 17:30 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-02

Prompt

We need to process raw clinical notes to perform medication reconciliation for an incoming patient. Which CLI command should be executed to extract named entities and output JSON?

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals2 of 4 complete

BenchmarkPass rate WITH skill on a verified run88

Live evalPass rate on real evaluated traces80%

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-40% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 2 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 50 days ago

Model	Method	Date	Lift
gemini-3.6-flash	verified	7/28/2026	+36%

Other measured skills in the registry, with their headline benchmark lift.