ansys-structural-workbench
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a comprehensive set of engineering and security quality gates. It utilizes standalone Python scripts (e.g., validate_job_spec.py, validate_results.py) to perform strict validation on all data ingested from or sent to the ANSYS environment, ensuring system integrity and preventing malformed inputs from influencing the execution environment.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes complex job specifications and geometry metadata, the risk of indirect prompt injection is mitigated by mandatory schema validation and the use of explicit delimiters (e.g., ANSYS_STRUCTURAL_JSON:) for structured data extraction. Instructions emphasize that the agent must rely on script-emitted evidence rather than qualitative judgment of potentially untrusted data.
- [DYNAMIC_EXECUTION]: The skill orchestrates live Mechanical sessions by generating Python snippets derived from local templates (mechanical_mesh_bridge.py). This dynamic execution is restricted to the ANSYS application context through a local MCP bridge and incorporates robust error handling and read-back verification to prevent unintended state mutations.
Audit Metadata