whatsapp-automation
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and replies to incoming WhatsApp messages, which creates a potential surface for indirect prompt injection where adversarial instructions in external messages could influence the agent.
- Ingestion points: Incoming messages handled in workflows such as 'Reply to Messages' in SKILL.md.
- Boundary markers: None explicitly specified to delimit or ignore instructions within external data.
- Capability inventory: The toolkit allows sending messages, media, and contacts.
- Sanitization: No specific filtering or sanitization of external content is described.
- [EXTERNAL_DOWNLOADS]: The skill setup requires configuring an external MCP server at
https://rube.app/mcp. This URL belongs to a well-known service providing the tool infrastructure for the skill's automation capabilities.
Audit Metadata