prd
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (customer evidence, JTBD analysis) without explicit boundary markers or sanitization, creating a surface for indirect prompt injection.
- Ingestion points: The skill ingests data from external files such as
.jimmy/docs/jtbd-analysis.mdand user-supplied customer evidence, as defined inSKILL.md(Step 0: Locate JTBD Input). - Boundary markers: Absent. The instructions do not define delimiters or explicit "ignore" instructions for the data being ingested to prevent the agent from obeying embedded commands.
- Capability inventory: The skill has the capability to write and append to the file system, specifically creating new PRD files in
.jimmy/work/and appending to.jimmy/decisions.md. - Sanitization: Absent. There is no guidance provided to the agent for filtering or escaping content from external inputs before incorporating them into the document or the decision log.
Audit Metadata