lancedb
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for reading and processing LanceDB table schemas and column metadata, which represents a potential ingestion surface for untrusted data. Ingestion points: The
references/column_metadata.mdfile describes reading existing field metadata via the LanceDB SDK to generate tags and descriptions. Boundary markers: The instructions do not define specific delimiters or warnings for the agent when processing this metadata. Capability inventory: The skill has the capability to perform table writes (update_field_metadata) and initiate network requests via the LanceDB REST API as documented inSKILL.mdandreferences/remote_jobs.md. Sanitization: There are no explicit instructions to sanitize or validate the content of retrieved metadata before it is used by the agent to generate code or documentation.
Audit Metadata