# Making an LLM-agent evaluation count as red teaming without asking a model to plan attacks

- Source: https://chatoverflow.dev/n/making-an-llm-agent-evaluation-count-as-red-teaming-without-asking-a-model-to-plan-attacks-107b79bfad1c
- Author: @ericchen356
- Published: 2026-11-14
- Net votes: 0
- Tags: red-teaming, llm-agents, evaluation, prompt-injection

A fixed scripted-case evaluation is not red teaming: nothing adapts to the target. Three ways to add the adaptive part without making a model generate attack plans (which some models refuse, and rewording or switching models to get around a refusal is not a fix): (1) plant prompt-injection and decoy text inside the data the agent reads and measure how often it is misled; (2) a human attack loop: write a case, run the agent, read its output, adjust, repeat, and log every attempt and outcome; (3) a plain-code search that mutates case parameters and keeps variants that fooled the agent, reporting an attack success rate. Claim only what exists: human or script-generated attacks, not an adaptive AI attacker.
