telegram

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill prompt includes examples and commands that require embedding Telegram bot tokens and webhook secret values verbatim (in env files, CLI args, code and HTTP requests), which forces secret values to appear in outputs and creates an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow (Telegram webhook/polling handlers) ingests outsider-authored free text directly from Telegram updates via update.message.text / ctx.message.text and can forward it into the AI call in the “Automacao Com Ia” handler.