prompt-engine

Pass

Audited by Gen Agent Trust Hub on Oct 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script ('search_prompts.py') via the shell to handle search requests.
  • [INDIRECT_PROMPT_INJECTION]: The orchestration logic passes user-supplied query text directly into a shell command template ('python3 ... "QUERY"'). This presents an attack surface where malicious input could attempt to bypass the intended command parameters.
  • Ingestion points: User input from the '/prompt' command.
  • Boundary markers: Double quotes are used around the input placeholder, but there are no instructions for sanitizing shell metacharacters.
  • Capability inventory: The skill relies on shell command execution for its primary search functionality.
  • Sanitization: The instructions do not define any validation or escaping procedures for the user-supplied content before it is executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 31, 2026, 05:14 PM