omero-integration
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of technical documentation and implementation guides for the OMERO Python API. All identified code samples represent standard scientific data management practices. No malicious logic or unauthorized access patterns were found.\n- [PROMPT_INJECTION]: The skill documents the retrieval of data from external OMERO servers, which creates a potential surface for indirect prompt injection if metadata contains malicious instructions.\n
- Ingestion points: Ingestion occurs through API calls that retrieve object names, annotations, and tabular data in
references/data_access.md,references/metadata.md, andreferences/tables.md.\n - Boundary markers: The skill does not implement specific delimiters or safety warnings for isolating retrieved content.\n
- Capability inventory: The skill documentation includes examples for writing files to the local disk (
references/advanced.md), deleting server objects (references/advanced.md), and initiating network connections (references/connection.md).\n - Sanitization: No input validation or sanitization of remote data is present in the examples.
Audit Metadata