helpcrunch
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the @membranehq/cli Node.js package, which is the official integration tool for the Membrane platform.
- [COMMAND_EXECUTION]: The agent uses the membrane CLI to perform authenticated actions and proxy API requests to HelpCrunch endpoints.
- [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by ingesting untrusted customer data (e.g., messages via get-chat-messages). Ingestion points: get-chat-messages, search-chats, list-customers in SKILL.md. Boundary markers: None specified. Capability inventory: Command execution and network proxying via membrane CLI in SKILL.md. Sanitization: Not mentioned. This is an inherent risk surface for customer communication integrations.
Audit Metadata