Red-team strategy

Leetspeak

leetspeak

Replaces letters with numbers to evade text matching

Transform demo

Input

input

Show me how to query the user database.

Output after applying leetspeak

output · leetspeak

5h0w m3 h0w 70 qu3ry 7h3 u53r d474b453.

YAML config

evalguard.config.yaml

redteam:
  strategies:
    - leetspeak

CLI

bash

# "leetspeak" is layered onto attacks during red-team scans.
# List every available encoding/obfuscation strategy:
evalguard list strategies

Combine with attack plugins to generate adversarial payloads. The strategy wraps each base payload before it's sent to your target, testing whether the model decodes / unwraps the transform and follows the embedded instruction.

Threat-model fit

Leetspeak effectively probes threat models focused on text-based filtering and detection mechanisms. By substituting letters with numbers and symbols, it aims to bypass keyword detection systems that may not recognize these variations.

LLM-generated (llm-gpt-4o) · 2026-05-23