whatsapp-agent

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill asks the agent to obtain API keys from the user and to embed them verbatim in curl commands, Authorization headers, and the Vercel REST payloads (ENV values), so the LLM would need to handle and output secret values directly.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.95). At runtime, the WhatsApp webhook handler (templates/app/api/whatsapp/webhook/route.ts) parses Kapso’s inbound payload and passes the outsider-authored message text (msg.text!.body) into handleMessage, which injects it into Gemini context via templates/lib/gemini.ts (history.push({ role: 'user', parts: [{ text }] })).

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)