nanobanana-mcp-image-generation

Warn

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation and execution of the nanobanana-mcp-server package using the uvx command. This package is an external dependency from an unverified developer.
  • [COMMAND_EXECUTION]: The MCP configuration relies on shell command execution (uvx) to launch the image generation server and manage environment variables.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by allowing the agent to read from and write to arbitrary file system paths.
  • Ingestion points: File paths provided to image_path, mask_path, file_path, and output_path parameters in the generate_image, edit_image, and upload_file tools.
  • Boundary markers: Absent. The skill does not provide instructions to the agent to delimit or ignore potentially malicious instructions embedded in input data or file paths.
  • Capability inventory: The skill possesses the capability to read local files (via upload_file and edit_image) and write files to the disk (via generate_image and edit_image).
  • Sanitization: Absent. No evidence of path validation or sanitization is provided in the skill instructions, which could allow for path traversal or unauthorized access to sensitive files like SSH keys or environment secrets.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 16, 2026, 10:43 PM
Security Audit — agent-trust-hub — nanobanana-mcp-image-generation