setting-up-a-custom-rest-source
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The workflow requires the agent to ingest and analyze data from external REST APIs to verify field mappings and data structures.
- Ingestion points: The
external-data-sources-preview-resourcetool fetches live data from arbitrary, user-definedbase_urlendpoints defined in the manifest. - Boundary markers: The instructions lack specific guidance on using delimiters or system-level instructions to ensure the agent ignores natural language instructions that might be embedded in the external API responses.
- Capability inventory: The agent possesses the capability to create permanent data sources via
data-warehouse-source-setupandexternal-data-sources-createbased on its interpretation of the external data. - Sanitization: While the manifest uses JSONPath for extraction, the agent is instructed to manually verify the rows, creating a window where malicious content in the response could influence the agent's behavior during the setup process.
Audit Metadata