resolution-aspect-decision
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of documentation, guidelines, and reference tables. It does not contain or invoke any scripts, executables, or code-heavy operations. All procedures are meant for a human operator to follow manually within a design environment.
- [SAFE]: No malicious patterns, prompt injections, or data exfiltration attempts were detected. The skill's behavior and instructions are consistent with its stated purpose of assisting in render planning and avoiding common mistakes like aspect ratio mismatches.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a target surface for user-provided delivery requirements, which could ingest untrusted data, but it lacks any exploitable capabilities. 1. Ingestion points: target_delivery_surfaces and framing_and_crop_strategy (contract.json). 2. Boundary markers: Present via the structured output response template (SKILL.md). 3. Capability inventory: None (the skill does not execute code, write files, or perform network operations). 4. Sanitization: Absent.
Audit Metadata