report-creator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its requirement to process external, untrusted data to define report structures. \n
- Ingestion points: Phase 1 of the orchestration model in
SKILL.mdinvolves gathering requirements from DevOps work items, mockups, screenshots, and user documents. \n - Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from obeying instructions that might be hidden within these requirement artifacts. \n
- Capability inventory: Across its workflow, the skill executes various
dxsCLI commands for project creation, data generation, and OData execution, and it writes output files to the local file system and/tmp. \n - Sanitization: The skill mitigates potential shell injection by mandating the use of the
--ops-fileargument when performing batch operations, ensuring that operation data is not directly parsed by the shell.
Audit Metadata