card-news-image-generator

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were found.
  • [EXTERNAL_DOWNLOADS]: The skill references and utilizes the nanobanana Model Context Protocol (MCP) for image generation. This is a functional tool dependency required for the skill's primary purpose.
  • [PROMPT_INJECTION]: An indirect prompt injection surface is present as the skill processes data from an external contents-manager output.
  • Ingestion points: Ingests the 'Visual Asset Plan' and 'Visual guidelines' from memory/files during the Auto-Load Protocol.
  • Boundary markers: Implements a 'Hard requirements checklist' in Step 1 to enforce 'no text' and 'no watermark' constraints.
  • Capability inventory: The skill is limited to calling the generate_image tool via MCP and performing file write operations to localized project directories.
  • Sanitization: Includes a dedicated 'Prompt Review' phase (Step 1) to validate and explicitly adjust prompts to ensure compliance with safety and style guidelines before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 01:00 AM