Diagnose, configure, generate, simulate, and report on NI Multisim circuits through the Multisim MCP server. Use when the user asks to install or configure Multisim MCP, generate a Multisim schematic, run circuit simulation, analyze waveforms, or produce an electronic design report.
Data & analysis
multisim-debug-circuit
Try it诊断并以最小、可回滚修改修复 Multisim 电路或仿真故障；use for invalid netlists, convergence failures, wrong waveforms, bias errors, or missing outputs
What it does
诊断并以最小、可回滚修改修复 Multisim 电路或仿真故障；use for invalid netlists, convergence failures, wrong waveforms, bias errors, or missing outputs
The skill document
调试 Multisim 电路
本 skill 引用 Multisim MCP 原始工具名；Harness 默认前缀为 mcp__multisim__。
- 保存问题描述、原始网表、分析命令、错误日志和已有实验 ID，不覆盖原始设计。
- 调用
runtime_status，把安装/COM/模板故障与电路故障分开。 - 对结构化设计先调用
diagnose_design；若已有同一设计的完成实验目录，把它作为experiment_dir附加。按顺序复核：网表语法、地节点、悬空节点、短路、极性、单位、 偏置、模型、分析命令、采样范围、饱和以及收敛条件。 - 若只有实验 ID，先用
get_experiment_summary，再按需读取log、netlist、commands或report；不要把二进制文件送入模型上下文。 - 提出能够解释故障的最小修改，明确修改前值、修改后值、理由和回滚方法。
- 把该最小修改写成明确
DesignPatch，调用evaluate_design_patch在同一硬性验收规范 下复测原设计和内存候选。若设计带权威source_netlist，只有确认需要时才显式允许 候选内存网表再生。不要把adoption_eligible当作提交授权。 - 检查前后诊断、两次实验和递归 manifest；失败或证据不足时保留“失败/未验证”。
- 只有用户另行批准时，才把通过候选交给本机审批绑定的持久化流程。
输出采用“症状—证据—根因—最小修改—复测结果—剩余风险”结构。不要在没有 复测数据时宣称问题已经解决。
English summary
Separate environment failures from circuit defects, inspect bounded evidence, apply the smallest reversible change, rerun into a new output directory, and compare measured results before declaring the issue fixed.
Related skills
将电路设计要求转成可计算判据并依据实验数据给出 PASS、FAIL 或未验证；use for acceptance testing, tolerance checks, and evidence-backed design review
根据工程要求生成安全、可复现的 Multisim 电路实验并核对理论与仿真结果；use when creating a new circuit, schematic, simulation, or experiment report
比较两个已注册 Multisim 实验的电路、设置、测量、验证和产物；use when evaluating a baseline against a candidate or explaining changed simulation results
基于已注册实验的真实产物撰写中英实验报告；use when producing a lab report, design record, reproducibility package, or formal HTML/PDF output
Diagnose hardware performance shortfalls with measurable problem definition, first-principles models, equation-led decomposition, 5M1E coverage, multidisciplinary review and evidence-ranked causal trees. Use for root-cause analysis, causal diagrams, out-of-spec parameters, 5 Whys, engineering contradictions or hardware troubleshooting.