Agent Skills › Cai-aa/CAE-Agent-Hub › aedt-evidence-electromagnetics

aedt-evidence-electromagnetics

GitHub

Ansys AEDT与PyAEDT电磁仿真自动化技能，支持Maxwell、HFSS等求解器。强调基于证据的执行流程，包括环境检查、脚本执行、参数审计及结果验证，确保仿真任务的准确性和可追溯性。

Skill/Ansys/aedt-evidence-electromagnetics/SKILL.md Cai-aa/CAE-Agent-Hub

触发场景

需要运行或验证Ansys Maxwell/HFSS/Q3D/Icepak电磁仿真任务使用PyAEDT进行自动化建模、求解设置或场数据提取需要生成包含完整证据链（日志、网格、结果文件）的仿真报告

安装

npx skills add Cai-aa/CAE-Agent-Hub --skill aedt-evidence-electromagnetics -g -y

更多选项

非标准路径

npx skills add https://github.com/Cai-aa/CAE-Agent-Hub/tree/main/Skill/Ansys/aedt-evidence-electromagnetics -g -y

不安装直接使用

npx skills use Cai-aa/CAE-Agent-Hub@aedt-evidence-electromagnetics

指定 Agent (Claude Code)

npx skills add Cai-aa/CAE-Agent-Hub --skill aedt-evidence-electromagnetics -a claude-code -g -y

安装 repo 全部 skill

npx skills add Cai-aa/CAE-Agent-Hub --all -g -y

预览 repo 内 skill

npx skills add Cai-aa/CAE-Agent-Hub --list

SKILL.md

Frontmatter

{
    "name": "aedt-evidence-electromagnetics",
    "description": "Evidence-first Ansys Electronics Desktop and PyAEDT automation for Maxwell, HFSS, Q3D, Icepak, and related electromagnetic workflows. Use when Codex must check AEDT\/PyAEDT, create or run a real AEDT design, verify the selected design and solution type, extract fields or matrices, export an .aedt project, or distinguish direct PyAEDT work from an AEDT MCP bridge."
}

AEDT Evidence Electromagnetics

Use the repository toolbox and a user-owned AEDT installation. Never infer a successful electromagnetic solve from ansysedt.exe, a project file, or a geometry screenshot.

Start

ai-cae-toolbox context-scope --solver aedt
ai-cae-toolbox bridge-plan --solver aedt --objective "<objective>"
ai-cae-toolbox create-run --solver aedt --case <case-name> --objective "<objective>"
ai-cae-toolbox write-smoke-template --solver aedt --run-dir <run-dir>

If MCP is available, prefer:

aedt_check_installation
aedt_run_pyaedt_script_file
scan_run_evidence
generate_run_report

Select the Correct Route

Use direct PyAEDT when the requested solver or API surface is not exposed by the installed MCP.
Use an AEDT MCP only for capabilities it actually exposes.
Do not build HFSS when the task requires Maxwell, or claim that an HFSS-focused MCP completed a Maxwell workflow.
Record AEDT version, PyAEDT version, project, design name, design type, solution type, and whether a new or existing desktop session was used.

Execute

Keep scripts in scripts/, source parameters in inputs/, project outputs in outputs/, and field plots/tables in exports/. Make every geometry dimension, material, excitation, boundary, setup, and sweep explicit.

For Maxwell matrix work, audit winding polarity and reciprocity. For HFSS, audit ports, modes, radiation boundaries, frequency setup, adaptive convergence, and S-parameter passivity. Read references/acceptance-gates.md before interpreting solver output.

Finish

Collect the AEDT project, solver/adaptive log, mesh statistics, numeric CSV/JSON, field images, and a report. Clearly label any external circuit calculation or Python-derived plot that is not a native AEDT result.

ai-cae-toolbox scan-evidence <run-dir>
ai-cae-toolbox generate-report <run-dir>

版本历史

fc7b93e 当前 2026-08-16 08:57

元信息

文件数: 0
版本: cb4d869
Hash: d77bbd9c
收录时间: 2026-08-16 08:57

aedt-evidence-electromagnetics