iblai-api-notification
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a well-documented API wrapper for the ibl.ai notification system. It uses standard environment variables for authentication and includes clear warnings for actions that have external effects or destructive potential.
- [INDIRECT_PROMPT_INJECTION]: The notification system ingests user-controlled data for message bodies and templates, creating a potential surface for indirect injection.
- Ingestion points: Data enters the system via the template_data and context fields in the builder's preview and send endpoints, as well as through CSV file uploads for recipient lists.
- Boundary markers: There are no specific instructions for the agent to use delimiters when interpolating user data, but the documentation focuses on API structure rather than LLM prompting.
- Capability inventory: The skill facilitates network communication with api.iblai.app to deliver notifications across email, push, and in-app channels.
- Sanitization: The references/templates.md file specifies that the system uses the bleach library to sanitize HTML templates and restricts the use of unauthorized Django template tags.
Audit Metadata