---name: regulatory-drafter description: Automates the drafting of regulatory documents (e.g., FDA CTD sections) with citation management and audit trails. keywords:

• regulatory-affairs
• fda
• ctd
• drafting
• compliance measurable_outcome: Produces a first-draft CTD section with >90% citation accuracy and reduced drafting time by 50%. license: MIT metadata: author: Artificial Intelligence Group version: "1.0.0" compatibility:
• system: Python 3.9+ allowed-tools:
• run_shell_command
• read_file ---"

Regulatory Drafter Skill

This skill assists regulatory affairs professionals by automatically generating sections of the Common Technical Document (CTD) from raw data and literature. It focuses on accuracy, traceability, and adherence to FDA/EMA guidelines.

When to Use This Skill

• Drafting CTD Sections: Automatically generate Module 2.4 (Nonclinical Overview) or Module 2.5 (Clinical Overview).
• Citation Management: When you need to synthesize multiple PDFs into a coherent narrative with inline citations.
• Audit Trails: When every claim must be traceable back to a source document.

Core Capabilities

1. Document Synthesis: Ingests clinical study reports (CSRs) and literature to write narrative text.
2. Citation Linking: Inserts hyperlinks or references to the source material.
3. Style Compliance: Adheres to eCTD formatting standards.

Workflow

1. Ingest Data: Point the agent to a folder of source PDFs/Data.
2. Draft: Run the drafter with a specific section target (e.g., "Nonclinical Overview").
3. Review: The agent outputs a Markdown/Word draft with annotations.

Example Usage

User: "Draft the Nonclinical Overview based on the toxicology reports in ./data/tox."

Agent Action:

python3 Skills/Anthropic_Health_Stack/regulatory_drafter.py --input "./data/tox" --section "2.4"