Skills
Navigate
skills/msapps-mobile/claude-plugins/whatsapp-send/Gen Agent Trust Hub

whatsapp-send

Fail

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download a repository from an untrusted GitHub account (github.com/lharries/whatsapp-mcp.git).
  • [COMMAND_EXECUTION]: Instructions include executing multiple shell commands for system monitoring and installation, such as ps, curl, and brew install.
  • [REMOTE_CODE_EXECUTION]: The setup workflow involves downloading remote code via git clone, compiling it with go build, and executing the resulting binary locally. This bypasses standard package verification and security review processes.
  • [PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by processing untrusted external data (WhatsApp messages and contact info) while maintaining file-sending and messaging capabilities. 1. Ingestion points: WhatsApp messages and contact search results. 2. Boundary markers: Absent. 3. Capability inventory: send_message, send_file, and send_audio_message tools. 4. Sanitization: Absent.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 28, 2026, 01:40 PM
Security Audit — agent-trust-hub — whatsapp-send