bbc-skill-tool
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a comprehensive technical guide for using the
builderbot_*toolset. It defines clear workflows for project creation, flow management, and bot deployment. - [PROMPT_INJECTION]: The skill architecture involves building AI assistants that process untrusted user input, establishing an indirect prompt injection surface. This is a design characteristic of the intended bot functionality.
- Ingestion points: WhatsApp user messages are processed by AI assistants (
add_chatpdf) as described inSKILL.mdandreferences/verticals.mdtemplates. - Boundary markers: The provided AI instruction templates lack explicit markers or instructions to isolate user input from system instructions.
- Capability inventory: The created bots possess network capabilities, including making HTTP requests (
add_http) and scraping URLs (scrapeUrl), as documented inreferences/advanced-patterns.md. - Sanitization: The templates do not include procedures for sanitizing or validating user input before it is used in subsequent automated actions.
Audit Metadata