bunnydoc

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the global installation of the @membranehq/cli package from npm. This is an official utility provided by the skill's author, Membrane, to manage authentication and proxy API requests securely.
  • [COMMAND_EXECUTION]: The skill uses the membrane CLI tool to execute various operations, including credential management, action discovery, and direct API requests to BunnyDoc.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from the BunnyDoc API (such as document records and team member lists) through commands like membrane action run and membrane request. The provided instructions lack explicit boundary markers or sanitization guidelines for the agent when handling this external content, which could potentially be used to influence the agent's logic if malicious data is present in the document management system.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 06:15 PM