table-creation
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and parse untrusted external data from user-provided CSV files and freeform text lists.
- Ingestion points: Processes data from pasted lists and CSV files in the 'Parse input data' workflow step.
- Boundary markers: No explicit delimiters or instructions are used to separate the parsed data from the agent's logic, allowing potential instructions within the data to influence the agent.
- Capability inventory: Orchestrates API calls for table creation, row uploads, and enrichment triggers via delegation to the
extruct-apiskill. - Sanitization: The skill performs domain normalization but does not describe validation or sanitization of arbitrary text fields provided in the input data.
Audit Metadata