Skip to the content
AI and Cybersecurity Learning and Practice Center Test automation Open the Learning Center

WeaknessASC-2026-0067Other weakness

Error messages and tool outputs that reveal internals

LowFixed

Verbose errors and tool results show file paths, schemas, prompts and keys to the model and to users, helping attackers plan the next step.

How to detect it

  • Stack traces or configuration in tool results
  • Internal URLs and schemas in responses

How to resolve it

  • Return generic errors to the model and user; log details server-side
  • Strip internal identifiers from tool outputs

How to investigate

  1. Review what internal details were exposed
  2. Rotate exposed secrets

Sources

Open it in the Learning Center's vulnerability database