prd
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by processing untrusted user data to generate persistent files.
- Ingestion points: The skill ingests feature descriptions provided by the user in Step 1 of the process (SKILL.md).
- Boundary markers: There are no explicit delimiters or specific instructions for the agent to treat the user input solely as data, which could lead the agent to follow instructions maliciously embedded in a feature description.
- Capability inventory: The skill has the capability to write files to the local file system in the
tasks/directory (SKILL.md). - Sanitization: No sanitization or validation steps are defined for the user-supplied feature descriptions or the resulting filenames before the content is saved.
Audit Metadata