pre-mortem

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted data from the local codebase to generate design recommendations.
  • Ingestion points: Codebase files are accessed and processed via subagents as instructed in SKILL.md.
  • Boundary markers: No specific delimiters or safety instructions are used to distinguish codebase content from agent instructions.
  • Capability inventory: The skill has the capability to write to the file system by saving design documents (e.g., {project_name}-design.md).
  • Sanitization: No sanitization, validation, or filtering of the ingested codebase content is performed before it is used in the plan generation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 05:32 PM