The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze data from external solver outputs, which could potentially contain malicious instructions.
Ingestion points: The agent reads solve.out (MAPDL logs), mesh metrics, and numeric result exports (CSV/JSON) as specified in SKILL.md and references/acceptance-gates.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within ingested logs were identified.
Capability inventory: The skill has the ability to execute shell commands via ai-cae-toolbox and RunWB2.exe.
Sanitization: No sanitization or validation of log content is specified before the agent processes the data.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for CAE orchestration.
Evidence: SKILL.md contains PowerShell blocks for ai-cae-toolbox (context-scope, bridge-plan, create-run, scan-evidence, generate-report) and RunWB2.exe for Workbench project orchestration.
[DYNAMIC_EXECUTION]: The skill utilizes scripting interfaces to manipulate the solver model.
Evidence: SKILL.md references the use of Mechanical Scripting/ACT with ExtAPI and PyMechanical gRPC for model-tree operations and session management.

ansys-mechanical-evidence-structural