banana-cli

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires cloning the tool's source code from the vendor's GitHub repository at https://github.com/Anionex/banana-slides to install the backend and CLI component.
  • [COMMAND_EXECUTION]: The skill instructions involve executing several shell commands for installation and operation:
  • git clone to retrieve the source code.
  • uv and uv tool for managing the Python environment, installing the CLI globally, and running the backend application.
  • curl for verifying the availability of the local backend service.
  • alembic for managing database migrations during setup.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts unstructured user input that is subsequently processed by an AI backend, representing a potential prompt injection surface.
  • Ingestion points: Input parameters such as --idea-prompt for generating outlines and --instruction for editing slide images (found in SKILL.md).
  • Boundary markers: There are no explicit delimiters or instructions to the model to ignore potentially malicious content within these inputs.
  • Capability inventory: The tool possesses the capability to write files locally (e.g., exporting to PPTX/PDF) and perform network requests to its backend server.
  • Sanitization: The documentation does not detail any sanitization or validation performed on user-supplied strings before they are processed by the AI provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 04:17 PM