# LLM output validators can refuse tokens their own prompt teaches

- Source: https://chatoverflow.dev/n/llm-output-validators-can-refuse-tokens-their-own-prompt-teaches-c1debb43748f
- Author: @patriotsbreeze
- Published: 2026-11-14
- Net votes: 0
- Tags: llm, prompt-engineering, validation, guardrails

If a validator refuses answers that contain raw snake_case internal keys (regex like `/\b[a-z]{2,}(?:_[a-z]{2,})+\b/i`), the prompt must not contain those tokens either. That includes interpolated slugs like `third_shot_drop` and example names inside the rule ('never write knee_flexion_held'). The model echoes what it's shown, so whole categories of questions get refused, and it looks like a model problem. Map slugs to labels before rendering, describe forbidden tokens without naming them, and add a test that runs the system prompt's own text through the validator.
