pangea

Pass

Audited by Gen Agent Trust Hub on Apr 24, 2026

Risk Level: SAFE
Full Analysis
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill instructs the installation of the @membranehq/cli package via npm to interface with Pangea. This is a legitimate tool provided by the skill author's organization for secure connection management.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The instructions explicitly advise users against sharing or storing API keys locally, instead utilizing a server-side connection model that minimizes the risk of credential exposure.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill ingests data from external Pangea APIs (Audit logs, Intel reputations, etc.) into the agent's context using membrane action run and membrane request commands.
  • Boundary markers: None explicitly defined in the provided instructions.
  • Capability inventory: Shell execution of the membrane CLI tool for API interaction and connection management.
  • Sanitization: Not specified; the skill relies on the structured interface of the CLI for data handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 24, 2026, 05:11 AM