security
AI Agent Router Security Auditor
Analyses agent routing patterns and validates security measures in multi-agent communication flows.
agent-security routing-validation communication-audit
prompt
# AI Agent Router Security Analysis You are an expert security auditor specialising in multi-agent systems. Analyse the provided agent routing configuration and communication patterns for security vulnerabilities. ## Agent Routing Configuration [paste your agent routing configuration here] ## Communication Logs (if available) [paste recent agent communication logs here] ## Analysis Requirements **1. Authentication & Authorisation** - Check if agents verify identity before message routing - Validate permission checks for cross-agent communication - Identify weak authentication mechanisms **2. Message Tampering Protection** - Analyse message integrity validation - Check for encryption in transit between agents - Review message signing and verification processes **3. Routing Logic Vulnerabilities** - Examine routing rules for injection attacks - Check for circular routing that could cause DoS - Validate agent discovery mechanisms **4. Input Sanitisation** - Review how messages are validated before routing - Check for command injection possibilities - Analyse parameter validation in routing decisions **5. Rate Limiting & Circuit Breaking** - Assess protection against message flooding - Check for proper backpressure mechanisms - Review timeout and retry logic ## Output Format Provide findings as: ### Critical Issues - **[Issue Name]**: Description, impact, and immediate fix ### Medium Priority Issues - **[Issue Name]**: Description and recommended remediation ### Security Recommendations - Specific improvements for the routing architecture - Best practices not currently implemented ### Implementation Checklist - [ ] Concrete security controls to implement - [ ] Monitoring and alerting improvements - [ ] Testing strategies for ongoing validation Focus on practical, actionable security improvements specific to this agent routing setup.
Use this when building or auditing multi-agent systems where security between agent communications is critical. Works with Claude, GPT-4, and Gemini to identify routing vulnerabilities and suggest specific security controls for your agent architecture.