openclaw-async-telegram-bot
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFE
Full Analysis
- [Data Exposure & Exfiltration] (SAFE): The script interacts with
api.telegram.orgto validate bot tokens. This is the legitimate endpoint for Telegram bot operations. Tokens are passed as arguments and are masked using a helper function before being printed to the console. - [Command Execution] (SAFE): The skill utilizes the
openclawCLI andcurl. All user-provided inputs used in commands are either validated against strict regex patterns (e.g.,^Nebutra[0-9]{3}_bot$) or sanitized (e.g., stripping non-alphanumeric characters for account IDs). Shell variables are consistently double-quoted to prevent injection. - [Indirect Prompt Injection] (SAFE): The skill processes JSON responses from the Telegram API. The data ingestion is limited to specific fields like
idandusername, which are then validated. No raw external data is directly interpolated into prompts or executed as code without validation. - [Metadata Poisoning] (SAFE): The metadata in
SKILL.mdaccurately describes the skill's functionality and does not contain deceptive instructions or hidden payloads.
Audit Metadata