create-knowledgebase
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from URLs and PDF files into a knowledge base for use with AI agents.
- Ingestion points: Data is accepted via the
--fileand--urlarguments inscripts/create_knowledgebase.py, as well as viacurlrequests documented inSKILL.md. - Boundary markers: There are no explicit markers or prompts within the ingestion script to distinguish between data and potential instructions in the source files.
- Capability inventory: The skill utilizes scripts for local file reading and network operations to the vendor's API at
https://api.bolna.ai. - Sanitization: Content is transmitted to the platform without client-side sanitization, relying on the receiving service or downstream agent for safety filtering.
Audit Metadata