write-prd
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted data from project source files to generate documentation.
- Ingestion points: The skill reads external content from
src/resources/,annotation-source.json, and.excalidrawfiles as specified inSKILL.md. - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious instructions embedded within these source files.
- Capability inventory: The skill is authorized to perform file writing operations to the
src/resources/directory to save the generated PRDs. It does not have access to sensitive execution tools or network operations. - Sanitization: There is no evidence of content sanitization or validation of the ingested prototype data before it is incorporated into the output.
- [SAFE]: The identified indirect prompt injection surface is considered safe in this context because the skill's capabilities are limited to writing non-executable Markdown documentation, which poses minimal risk to the host environment or user data.
Audit Metadata