tg-rich-messages
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate functionality for interacting with the Telegram Bot API as described in the documentation. No malicious code, obfuscation, or persistence mechanisms were detected.
- [DATA_EXFILTRATION]: The skill communicates with api.telegram.org, which is the official and expected endpoint for Telegram bot operations. It correctly instructs users to provide sensitive credentials via environment variables (TELEGRAM_BOT_TOKEN), avoiding the use of hardcoded secrets.
- [EXTERNAL_DOWNLOADS]: Code examples utilize standard Python libraries and the httpx package for API interaction. These are established and appropriate tools for the described network communication tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a risk surface for indirect prompt injection when formatting data for Telegram. Ingestion points: content is processed through the rich_message field in SKILL.md and function arguments in rich_ui.py. Boundary markers: the skip_entity_detection flag is available to prevent automatic link parsing by the client. Capability inventory: network requests are limited to the Telegram API via urllib and httpx. Sanitization: the rich_ui.py helper validates text length and block nesting while treating inputs as literal strings to prevent formatting-based attacks.
Audit Metadata