punk-avatar

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data, including text descriptions and images, which are then interpolated into a template to generate a final prompt for an image generation tool.
  • Ingestion points: User-supplied images and text descriptions are analyzed and processed in SKILL.md under the 'Workflow' section (Step 1).
  • Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands when merging user-provided traits into the references/avatar-prompt-blueprint.md structure.
  • Capability inventory: The skill triggers image generation via an internal tool and performs file-write operations to the punk-assets/ directory as described in SKILL.md (Step 7).
  • Sanitization: There are no documented sanitization, escaping, or filtering steps applied to the user-derived subject fields before they are used to compile the final image-generation brief.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 04:38 AM
Security Audit — agent-trust-hub — punk-avatar