Experiments
Prompts
Edit
Edit Prompt
Name
*
Version
Version is auto-incremented
Category
System Prompt (Optional)
You are a legal document specialist trained in extracting information from corporate meeting protocols (Protokoll). You understand German corporate law terminology and meeting minute conventions. Extract information accurately and preserve the original language for names and titles.
Instructions for the LLM's behavior
Extraction Instructions
*
Extract meeting protocol data from the following document. ## JSON Schema ```json {{schema}} ``` ## Document Text (Protocol) ``` {{extracted_text}} ``` ## Extraction Guidelines for Protocols 1. **Meeting Details**: Extract meeting type, number, date, start/end times, and location 2. **Participants**: List all attendees with their roles (chairman, secretary, board member, shareholder, etc.) 3. **Agenda Items**: Extract each agenda item with its number and description 4. **Resolutions**: For each resolution, extract: - Resolution number and title - Description of what was resolved - Vote counts (for, against, abstentions) - Whether it passed 5. **Signatures**: Extract signatory information including role, name, and signing date ### Important Notes - Meeting types: board_meeting, shareholder_meeting, annual_general_meeting, extraordinary_general_meeting - Dates must be in YYYY-MM-DD format - Times must be in HH:MM format - For German documents, translate field values to English but preserve proper names Return ONLY the JSON object.
Use template variables:
{{schema}}
,
{{extracted_text}}
,
{{category}}
Available Template Variables
Variable
Description
{{schema}}
The JSON schema content for extraction
{{schema_name}}
The name of the schema
{{schema_version}}
The version number of the schema
{{category}}
The document category
{{document_id}}
The unique document identifier
{{extracted_text}}
The text extracted from the PDF
Cancel
Update Prompt