analytics
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it ingests and processes untrusted data from external sources such as API responses and data warehouse results.
- Ingestion points: The agent ingests external data from GrowthBook API endpoints and SQL query results (CSV/JSON) across all referenced workflows, notably in
references/analytics-explore.mdandreferences/sql-query.md. - Boundary markers: The instructions do not define explicit boundary markers or delimiters when presenting retrieved data to the agent, nor do they include specific instructions to ignore potential commands embedded within the data.
- Capability inventory: The skill has the capability to perform write operations (POST/PUT/DELETE) to the GrowthBook environment and execute shell-based API calls via the
gb-calltool. - Sanitization: No explicit sanitization or filtering of warehouse data or API response content is mentioned before the agent interprets the results.
Audit Metadata