prompt-enhance

Warn

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a shell command python3 {PROMPT_ENGINE_DIR}/scripts/search_prompts.py "KEY_TERMS" --limit 5 where KEY_TERMS are derived from the user's input prompt. The lack of sanitization for these terms allows for shell injection attacks, where a user can provide input containing characters like ;, &, or | to run unintended commands on the host environment.\n- [PROMPT_INJECTION]: The skill processes untrusted user prompts to generate search terms and enhanced prompt output, creating a surface for indirect prompt injection.\n
  • Ingestion points: User input prompts provided to the enhancement workflow in SKILL.md.\n
  • Boundary markers: Absent. The skill does not employ delimiters or specific instructions to treat user input as non-executable data.\n
  • Capability inventory: Subprocess execution via python3 in SKILL.md.\n
  • Sanitization: Absent. There is no evidence of input validation, filtering, or escaping before the user-provided data is used to construct shell commands.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 22, 2026, 09:14 PM