Core Feature

Trust in ConnectOnion

Flexible, bidirectional trust configuration for agent interactions

Read: Why we chose "trust" as our keyword→

Quick Start

trust_config.py

from connectonion import Agent, need

# Simple trust levels
translator = need("translate", trust="strict")   # Production: verified only
analyzer = need("analyze", trust="tested")       # Default: test first
scraper = need("scrape", trust="open")          # Development: trust all

# For your own agent
agent = Agent(
    name="my_service",
    tools=[process_data],
    trust="strict"  # Who can use my services
)

Python REPL

Interactive

>>> translator = need("translate", trust="strict")

>>> print(translator)

<Agent: translate (strict trust)>

>>> agent = Agent(name="my_service", tools=[process_data], trust="strict")

>>> print(f"Agent {agent.name} configured with strict trust policy")

Three Forms of Trust

1. Trust Levels (String)

Simple predefined levels for common scenarios:

open→Trust everyone (development)

tested→Test before trusting (default)

strict→Verified agents only (production)

2. Trust Rules (Dict)

Fine-grained control with custom rules:

custom_trust.py

trust_rules = {
    "allow": ["agent1", "agent2"],     # Whitelist
    "deny": ["untrusted_agent"],       # Blacklist  
    "require_auth": True,               # Authentication
    "test_timeout": 30,                 # Test duration
    "max_interactions": 100             # Rate limiting
}

agent = Agent("secure", trust=trust_rules)

3. Trust Functions

Dynamic trust evaluation with custom logic:

dynamic_trust.py

def custom_trust(agent_name: str, context: dict) -> bool:
    """Evaluate trust based on custom logic"""
    if context.get("environment") == "production":
        return agent_name in VERIFIED_AGENTS
    return True  # Open in development

agent = Agent("dynamic", trust=custom_trust)

Bidirectional Trust

Trust works both ways in ConnectOnion:

Outbound Trust

Who I trust to use

need("service", trust="...")

Inbound Trust

Who can use me

Agent(..., trust="...")

Choose Your Trust Level

Development?

Use open

Testing/Staging?

Use tested

Production?

Use strict

Common Patterns

Progressive trust elevation
Environment-based trust
Mutual authentication
Rate limiting & quotas

Security Note

Always use strict trust in production environments to prevent unauthorized access.

LLM Function

@xray Decorator