supply-chain-analytics
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill includes a Python function
extract_from_apiin the ETL section that uses therequestslibrary to fetch data from an external endpoint (https://api.wms.com/v1/inventory). This is a legitimate feature for data analysis skills but involves network operations. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external SQL databases and REST APIs (as shown in the
SupplyChainETLclass). While the provided code focuses on structured data processing, this mechanism creates a surface for indirect prompt injection if the source data contains untrusted natural language content that is subsequently presented to the agent. - Ingestion points: External SQL databases and REST APIs via
extract_from_sqlandextract_from_apifunctions inSKILL.md. - Boundary markers: None identified in the provided templates.
- Capability inventory: The skill uses
sqlalchemyfor database operations (read/write) andrequestsfor network data fetching. - Sanitization: The skill includes basic data cleaning (duplicates, types) but lacks specific sanitization for preventing prompt injection from ingested strings.
Audit Metadata