statsmodels
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources via
sandbase_run, which could potentially contain malicious instructions embedded in datasets. - Ingestion points: External data is fetched using
sandbase_runandsandbase_run_getas described inSKILL.mdandreferences/sandbase-api-map.md. - Boundary markers: The skill does not define specific delimiters for external content but instructs the agent to inspect shapes, types, and units before transformation.
- Capability inventory: The skill involves local file access, implementation of statsmodels logic, and network tool calls to the vendor's infrastructure.
- Sanitization: The instructions emphasize validation against baselines and reporting diagnostics, which mitigates unexpected results from poisoned data.
Audit Metadata