Skip to main content

Red Team Troubleshooting Guide

Common issues encountered when red teaming LLM applications with promptfoo.

IssueDescription
Attack GenerationConfiguration errors in system scope, permissions, or available actions can prevent effective attack generation.
Connection ProblemsAuthentication failures, rate limiting issues, and incorrect endpoint configuration can prevent successful API connections.
False PositivesInsufficient system context in grader configuration can lead to incorrect vulnerability assessments.
Multi-Turn SessionsSession management issues can disrupt conversation context in both client and server-side implementations.
Multiple Response TypesResponse parsing errors occur when handling non-standard formats, guardrails, or error states.
Remote GenerationCorporate firewalls may block access to remote generation endpoints due to security policies around adversarial content.