whatsapp-integration
Pass
Audited by Gen Agent Trust Hub on Nov 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Installs the
@membranehq/clipackage from the NPM registry to enable interaction with the Membrane platform. - [COMMAND_EXECUTION]: Utilizes the
membraneCLI to perform authentication, discovery, and execution of WhatsApp-related actions and raw API requests. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (WhatsApp messages and contact data) which presents a surface for indirect prompt injection.
- Ingestion points: WhatsApp chat messages, business profiles, and contact cards retrieved via the CLI commands.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are provided in the documentation for the agent to use when processing text from WhatsApp.
- Capability inventory: Includes capabilities to send messages, reactions, and various media types, as well as making arbitrary proxy requests to the WhatsApp API via
membrane request. - Sanitization: No explicit sanitization or filtering of incoming WhatsApp content is described in the skill instructions.
Audit Metadata