nanobanana-ppt

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill description mentions 'document analysis', which indicates an attack surface for indirect prompt injection where instructions hidden in processed documents could attempt to influence the agent's behavior.
  • Ingestion points: Documents processed for deck generation as described in SKILL.md.
  • Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are defined in the manifest.
  • Capability inventory: The manifest does not include internal scripts; capabilities are inherited from the agent's environment.
  • Sanitization: No sanitization or filtering of external document content is specified.
  • [NO_CODE]: The analyzed skill contains only metadata and usage instructions, with no executable scripts or logic included in the package.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 03:58 PM
Security Audit — agent-trust-hub — nanobanana-ppt