vector-databases
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes well-known, legitimate Python packages for vector database interactions and follows industry-standard security practices for credential management by referencing environment variables instead of hardcoding secrets.
- [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for indexing and retrieving external content, which inherently introduces an attack surface for indirect prompt injection if malicious data is retrieved into the agent's context.
- Ingestion points: External document content is ingested via the
add_documentsmethod inscripts/vector_store_manager.py. - Boundary markers: The skill does not implement delimiters or safety warnings for retrieved content within its storage logic.
- Capability inventory: The skill interacts with database APIs and local storage for vector persistence (scripts/vector_store_manager.py), but does not possess dangerous capabilities like arbitrary code execution or shell access.
- Sanitization: No sanitization or validation of the document content is performed before it is indexed or retrieved.
Audit Metadata