palantir-migration-deep-dive
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion pipeline that reads from external SQL databases.
- Ingestion points: The skill reads data from external sources using
pandas.read_sqlin Steps 2, 3, and 5 withinSKILL.md. - Boundary markers: The skill does not implement specific boundary markers or instruction-delimiting warnings for the data it processes.
- Capability inventory: The skill possesses
Writeaccess to datasets and execution capabilities for Python scripts, allowing it to move and transform data. - Sanitization: There is no explicit sanitization of the external data content before it is processed by the migration scripts, which is a standard characteristic of data migration tools but represents an inherent injection surface.
Audit Metadata