setting-up-warehouse-properties
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from external warehouse tables and materialized views into the agent context and PostHog environment.
- Ingestion points: The skill uses
external-data-schemas-retrieveandview-getto read schema and column information from external sources as described in SKILL.md. - Boundary markers: There are no explicit instructions to the agent to use boundary markers or delimiters when handling the ingested values, although the mapping is structural.
- Capability inventory: The skill utilizes
custom-property-sources-createto write the retrieved warehouse data into person or group properties. - Sanitization: The skill instructions include a proactive safety measure to exclude sensitive identity properties and security-critical data like password hashes and tokens from the sync proposal.
Audit Metadata