Repository files navigation

Integrates 🦜Parlant's ARQ (Attentive Reasoning Queries) into 🔷Microsoft Agent Framework via middleware.

• Guideline Matching: 🚦Semantic matching with embeddings (cached in.embedding_cache.pkl)
• Journey State Tracking: Conversational flow as state machines
• ARQ Validation: 🛡️ Anchor → Reason → Predict (3-phase validation)
• Context Propagation: Shared state across agent handoffs
• Compliance: Validate responses before delivery ✔️

uv sync

fromcore.modelsimportGuideline,Journey,JourneyState,StateTypefrommiddleware.query_gatewayimportARQGatewayMiddlewarefromagent_framework.azureimportAzureOpenAIChatClient# Define guidelinesguidelines= [Guideline(id="g1",condition="Customer frustrated",action="Acknowledge feelings",priority=10)]# Create agent with ARQ middlewareagent=chat_client.create_agent(name="Agent",instructions="Your instructions",middleware=[ARQGatewayMiddleware(guidelines=guidelines)])# Runresult=awaitagent.run("User message")

View Embedding Cache:

python view_cache.py# Shows .embedding_cache.pkl contents

• demo_simple_arq_validation.py - Basic ARQ validation
• demo_journey_state_tracking.py - Journey state flow
• demo_customer_support_arq.py - Multi-agent support with handoffs
• demo_travel_booking_arq.py - Travel booking workflow (paper example)

ARQ Flow: Guideline Matcher → Journey Manager → ARQ Generator → Agent Executor
Middleware: Non-invasive integration viaAgentMiddleware
Cache: Embeddings stored in.embedding_cache.pkl (persistent)

ARQ (Attentive Reasoning Queries) are prompts designed to guide an AI to think step-by-step, focus on key details, and produce accurate, logically grounded answers by directing its attention.

• ARQ Paper - Attentive Reasoning Queries
• Parlant - Original ARQ implementation
• Microsoft Agent Framework