front

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli global package from NPM, which is a vendor-owned resource for the Membrane platform.
  • [COMMAND_EXECUTION]: The skill executes membrane CLI commands for authentication and API interaction, which are legitimate functions for this integration.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection attack surface through ingested Front data.
  • Ingestion points: Conversations, contacts, and custom API data retrieved via membrane action run and membrane request in SKILL.md.
  • Boundary markers: None identified.
  • Capability inventory: Ability to modify or delete resources and perform arbitrary requests using the membrane CLI in SKILL.md.
  • Sanitization: No explicit validation or sanitization steps are defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 05:18 PM