AutoGen Integration

Add trust verification to AutoGen multi-agent conversations. Ensure only verified agents participate.

Installation

pip install nerq-autogen

Quick Start

Wrap your AutoGen agents with trust verification. The wrapper checks each agent's trust score before allowing it to participate in conversations.

# 1. Import the trust wrapper
from nerq_autogen import TrustVerifiedAgent

# 2. Create a trust-verified agent
import autogen

config_list = autogen.config_list_from_json("OAI_CONFIG_LIST")

agent = TrustVerifiedAgent(
    name="researcher",
    agent_id="researcher-agent-v2",
    nerq_api_key="your-nerq-api-key",
    min_trust_score=60,
    llm_config={"config_list": config_list},
    system_message="You are a research assistant."
)

# 3. Use in a conversation
user_proxy = autogen.UserProxyAgent(
    name="user",
    human_input_mode="NEVER"
)

user_proxy.initiate_chat(
    agent,
    message="Summarize the latest AI safety research."
)

Multi-Agent Conversations

Verify all agents in a group chat before the conversation starts.

from nerq_autogen import TrustVerifiedGroupChat

# Create multiple verified agents
researcher = TrustVerifiedAgent(
    name="researcher",
    agent_id="researcher-v2",
    nerq_api_key="your-key",
    min_trust_score=60,
    llm_config={"config_list": config_list}
)

writer = TrustVerifiedAgent(
    name="writer",
    agent_id="writer-v3",
    nerq_api_key="your-key",
    min_trust_score=70,   # higher threshold for writing agents
    llm_config={"config_list": config_list}
)

# Group chat with trust verification
group_chat = TrustVerifiedGroupChat(
    agents=[researcher, writer, user_proxy],
    messages=[],
    max_round=10
)

manager = autogen.GroupChatManager(groupchat=group_chat)
user_proxy.initiate_chat(manager, message="Write a report on AI trends.")

Configuration

FAQ