telegram-bot-builder
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides implementation patterns for bots that ingest and process untrusted text messages from Telegram users.
- Ingestion points: Untrusted data enters the context via
bot.on('text', ...)handlers described inreferences/patterns.md. - Boundary markers: The provided code snippets do not include explicit boundary markers or warnings to ignore instructions embedded within user messages.
- Capability inventory: The skill patterns include capabilities for network operations (Telegram API calls via
ctx.replyandbot.telegram.sendMessage) and persistence operations (database writes via Redis/SQL inreferences/sharp_edges.md). - Sanitization: There is no mention of sanitizing or escaping user-provided text before it is processed or used in downstream logic.
Audit Metadata