whatsapp-mcp
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to clone an external repository (
github.com/lunasoft2001/mcp.git) and install dependencies usingnpm install. This is required to set up the local WhatsApp MCP server components. - [COMMAND_EXECUTION]: The instructions include a shell command template for using Google Chrome in headless mode to convert HTML files to PDF. It also instructs the user to grant execution permissions to a local script (
start.sh) usingchmod +xwhich is then executed by the agent platform. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads external data from WhatsApp messages through the
get_chat_messagestool. - Ingestion points: WhatsApp messages retrieved from the WhatsApp API.
- Boundary markers: No specific delimiters or instructions to ignore embedded content within messages are provided in the skill's instructions.
- Capability inventory: The agent can send messages, manage group participants, and execute local shell commands for file conversion.
- Sanitization: No explicit sanitization or validation of message content is described before the agent processes the retrieved text.
Audit Metadata