telegram

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill includes a dedicated 'Security' section in SKILL.md that explicitly warns the agent to treat Telegram message content as data and never as instructions. This is a best-practice defense against indirect prompt injection.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the installation of the telegram-agent CLI tool via standard package managers (npm, pnpm, bun) from official registries. This is standard for distributing CLI-based skills.
  • [DATA_EXFILTRATION]: While the skill manages highly sensitive Telegram session tokens and message content, it provides clear instructions on data boundaries and warns against following untrusted URLs or constructing API calls from message content.
  • [COMMAND_EXECUTION]: All functionality is mediated through the telegram-agent CLI. The skill provides clear error handling and usage patterns that emphasize safe data processing using jq.
  • [CREDENTIALS_UNSAFE]: The skill correctly instructs users to manage sensitive credentials (TELEGRAM_API_ID, TELEGRAM_API_HASH) using environment variables and local session files, avoiding hardcoded secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 11:22 AM
Security Audit — agent-trust-hub — telegram