Skip to content

Strict Referenced Agent

Strict-Referenced Medical AI Assistant. This endpoint provides a strict-referenced agent that: self-evaluation and retry logic, adding ~3-4 seconds to response time.

  • Try it: /agents
  • API Endpoint: /agent/strict-referenced
  • FastAPI Docs: /api-docs

Key Features

  • Requires at least 5 searches before providing any answer
  • Cannot use its own knowledge - everything must be confirmed through research tools
  • All responses must have full references
  • Will refuse to answer if insufficient sources are found

Request Parameters

Parameter Description
messages List of chat messages with role and content
profile Patient profile information (name, location, language, category, history)
model Optional OpenAI model selection (default: gpt-4o-mini)
with_evaluation Optional boolean (default: false). If true, includes

System Prompt

System prompt not found in agent source code.

User Prompt Template

AGENT_USER_PROMPT_TEMPLATE

Profile: {profile}

Messages:
{messages}

API Reference

Endpoint: POST /agent/strict-referenced

Interactive Docs: FastAPI Swagger UI

Example Usage

curl -X POST http://localhost:8000/agent/strict-referenced \
  -H 'Content-Type: application/json' \
  -d '{
    "messages": [{"role": "user", "content": "What are the symptoms of diabetes?"}],
    "profile": "Patient with family history of diabetes"
  }'