not-a-vibe-coder

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user input to generate project planning documents (such as PRD.md and TechSpec.md) which are then treated as the agent's persistent memory. This creates a surface where malicious instructions could be embedded in the documentation and influence the agent's behavior when the files are re-read in subsequent sessions. \n
  • Ingestion points: User input during documentation drafting in Phases 1, 2, and 2.5 of SKILL.md. \n
  • Boundary markers: None present; the skill lacks delimiters or explicit instructions to ignore potentially embedded commands within the generated files. \n
  • Capability inventory: The skill performs file system read and write operations across the project directory. \n
  • Sanitization: No sanitization or validation of user-provided content is performed before it is persisted to the planning documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 01:28 AM
Security Audit — agent-trust-hub — not-a-vibe-coder