dropbox-expert

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation provides instructions to download the dbxcli utility from the official Dropbox GitHub repository and the Dropbox-Uploader script from a widely recognized community repository on GitHub. Both sources are established and appropriate for the skill's functionality.
  • [COMMAND_EXECUTION]: Installation procedures for the documented CLI tools involve standard operations like chmod for setting executable permissions and sudo mv for moving binaries to system-wide paths. These are routine installation steps for developer-oriented tools.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8). Evidence Chain: (1) Ingestion points: The skill retrieves untrusted metadata such as file and folder names from external Dropbox accounts via scripts/files.py and references/files-api.md. (2) Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore potentially malicious content embedded in file names. (3) Capability inventory: The skill provides extensive file system operations including write, delete, and sharing actions. (4) Sanitization: No specific escaping or validation of external metadata is implemented in the provided Python scripts. Given the intended use-case of a file management utility, this surface is noted but does not escalate the verdict.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 11:27 AM
Security Audit — agent-trust-hub — dropbox-expert