The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access attempts were found in the skill's instructions or configuration files.
[COMMAND_EXECUTION]: The skill uses a domain-specific command-line interface, ai-cae-toolbox, to manage simulation contexts and objectives. This execution is confined to the specific engineering automation tasks described in the metadata.
[DYNAMIC_EXECUTION]: The tool aedt_run_pyaedt_script_file is used to execute PyAEDT scripts. This is a standard and expected capability for the intended primary purpose of automating Ansys Electronics Desktop workflows.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it reads external solver logs and simulation data.
Ingestion points: The agent reads solver logs, adaptive logs, and numeric data from the simulation run directory as specified in SKILL.md.
Boundary markers: The instructions in references/acceptance-gates.md serve as a validation framework, directing the agent to look for specific engineering signals (e.g., convergence history, reciprocity) rather than interpreting free-form text as commands.
Capability inventory: The skill has the capability to execute shell commands via ai-cae-toolbox and run Python automation scripts.
Sanitization: The skill relies on 'acceptance gates' to validate the physical and technical integrity of the simulation results before finalizing reports.

aedt-evidence-electromagnetics