dask
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external files and data ("Inspect the available files, runtime, versions, inputs"), which constitutes an ingestion surface for potential indirect prompt injection attacks where malicious data could attempt to influence the agent's behavior.
- Ingestion points: Workflow Step 1 in
SKILL.mdinstructs the agent to inspect files, runtime, and inputs. - Boundary markers: The skill explicitly instructs the agent to "Preserve the user's chosen stack, source material, and authorization boundaries" and to restate constraints before proceeding.
- Capability inventory: The skill utilizes the
sandbase_runtool for external operations and guides the generation of Dask code for local data processing. - Sanitization: The instructions include "Quality gates" to inspect data shapes, types, units, and missing values, which serves as a form of validation for processed content.
Audit Metadata