user-story
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input to generate structured user stories and acceptance criteria. This input is also intended to be passed as arguments to a local Python helper script. The absence of boundary markers or sanitization logic constitutes a standard injection surface.
- Ingestion points: User input provided via the
argument-hint('[feature or user need]') and theARGUMENTS:line inSKILL.md. - Boundary markers: The instructions do not define clear delimiters to separate user data from instructions.
- Capability inventory: The skill involves executing a local Python script (
scripts/user-story-template.py) via the shell using arguments derived from user input. - Sanitization: Neither the prompt instructions nor the Python script perform validation, escaping, or filtering of the user-supplied content before processing.
Audit Metadata