campaign-cleaner

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill guides the installation of the official Membrane CLI tool (@membranehq/cli) via npm to facilitate interaction with the platform.
  • [COMMAND_EXECUTION]: Makes extensive use of the membrane command-line interface for managing user authentication, service connections, and running business logic actions.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: Adheres to security best practices by instructing the agent to never request raw API keys or tokens, instead delegating all credential management to the Membrane platform.
  • [DYNAMIC_EXECUTION]: Supports the dynamic creation of integration actions through the membrane action create command, which generates new executable logic based on natural language descriptions.
  • [INDIRECT_PROMPT_INJECTION]: Involves a potential surface for indirect injection by passing natural language intents to the CLI for action discovery, requiring the agent to remain vigilant for instructions in external metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 04:49 PM
Security Audit — agent-trust-hub — campaign-cleaner