telegram-alerts-hardened
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements security best practices for handling sensitive credentials (TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID) by requiring them to be stored in an .env file and accessed via environment variables rather than being hardcoded or displayed in the UI.- [SAFE]: Explicit guardrails are provided to prevent data exfiltration, specifically forbidding the transmission of local file contents, system information, or non-trading data through the Telegram bot API.- [SAFE]: The instructions include specific measures to prevent command injection, such as requiring the agent to construct only straightforward sendMessage API calls and prohibiting the use of shell expansion, eval, or piped input.- [SAFE]: A destination integrity guardrail ensures that alerts are only sent to the pre-configured chat ID from the .env file, preventing the redirection of sensitive financial data to unintended or attacker-controlled Telegram destinations.- [SAFE]: The skill requires a mandatory user confirmation step before sending the first alert in a session, providing a human-in-the-loop check to verify configuration before any data leaves the local environment.
Audit Metadata