channel-bot
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of documentation and development guidelines for a messaging backend. It does not contain executable scripts or automation code that could be exploited.
- [INDIRECT_PROMPT_INJECTION]: The skill documents the entry points for external user messages from Telegram and Slack into the AI agent pipeline.
- Ingestion points: Inbound messages processed by platform-specific adapters in
services/channels/(SKILL.md). - Boundary markers: Not explicitly defined in this documentation, though it refers to the standard agent pipeline context.
- Capability inventory: Inbound messages trigger agent runs and response generation, which may include rendering chart outputs via
chart_render.py. - Sanitization: The instructions strictly mandate HMAC/Signing Secret verification for all webhooks to ensure the integrity of the data source before it is processed.
Audit Metadata