What document formats does this skill support?

The skill can process clinical trial Protocols and Statistical Analysis Plans in PDF, DOCX, and plain text formats.

What happens if a Protocol and SAP conflict?

The skill performs cross-referencing; it typically gives precedence to the SAP for analysis decisions and the Protocol for study design, while flagging discrepancies for human review.

Can it handle very large clinical documents?

Yes, it uses a targeted reading strategy to navigate documents over 20 pages, focusing on key sections like the synopsis, table of contents, and specific analysis chapters.

Is the output JSON ready for production use?

The output is designed as a structured contract for downstream tools, though it includes 'reviewer_attention' fields to ensure a biostatistician can verify any ambiguous interpretations.

Trial Metadata Extractor

Name: Trial Metadata Extractor
Author: Appsilon

byAppsilon

•

数据科学与机器学习

Extracts structured metadata from clinical trial Protocol and Statistical Analysis Plan (SAP) documents into standardized JSON for pharma automation.

The Trial Metadata Extractor skill automates the complex task of parsing clinical trial documents to create a structured 'metadata contract' for downstream automation. By analyzing Protocols and Statistical Analysis Plans (SAPs), it identifies critical study components including design elements, primary and secondary endpoints, population criteria, and visit schedules. It prioritizes regulatory fidelity by preserving exact source wording and uses intelligent cross-referencing to resolve discrepancies between documents, providing a high-quality foundation for ADaM planning and TLG shell generation.

主要功能

01Automated identification of gaps and items requiring reviewer attention

0222 GitHub stars

03Structured JSON output for ADaM and TLG automation pipelines

04Intelligent chunked reading for large PDF clinical documents

05Cross-document validation between Protocol and SAP documents

06Fidelity-focused extraction preserving regulatory-grade definitions

使用场景

01Summarizing complex study designs and analysis plans for biostatistical review

02Preparing structured inputs for automated mock TLG shell generation

03Mapping clinical endpoints to statistical analysis methods across multiple documents

主要功能

01Automated identification of gaps and items requiring reviewer attention

0222 GitHub stars

03Structured JSON output for ADaM and TLG automation pipelines

04Intelligent chunked reading for large PDF clinical documents

05Cross-document validation between Protocol and SAP documents

06Fidelity-focused extraction preserving regulatory-grade definitions

使用场景

01Summarizing complex study designs and analysis plans for biostatistical review

02Preparing structured inputs for automated mock TLG shell generation

03Mapping clinical endpoints to statistical analysis methods across multiple documents