012-agile-epic

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to run the system date command to retrieve the current date for documentation headers. This is a legitimate and benign use of command execution.
  • [PROMPT_INJECTION]: The skill processes user input to populate an epic template, which creates an indirect prompt injection surface. * Ingestion points: User answers to the 12 questions defined in references/012-agile-epic.md. * Boundary markers: No delimiters are used to separate user data from the template structure. * Capability inventory: The agent has access to terminal execution (date), but there is no mechanism for user input to flow into this command or other dangerous sinks. * Sanitization: No input validation or escaping is applied to the user-provided text.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 06:06 PM