prompt-injection

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill contains phrases such as "Ignore previous instructions" and "What is your system prompt?". These are explicitly presented within the context of 'Step 5: Check for Prompt Leaking' as examples of common extraction attempts for auditors to test. They are not instructions directed at the agent itself.
  • [COMMAND_EXECUTION]: The YAML frontmatter includes 'Bash' and 'Write' in the 'allowed-tools' list. These permissions are scoped to the skill's purpose of auditing codebases (e.g., using 'Grep' to find API calls) and documenting findings in report format.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 05:19 PM
Security Audit — agent-trust-hub — prompt-injection