The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local SPICE simulation binaries (ngspice, LTspice, or Xyce) when the --spice-enhanced flag is provided. This is a core functional requirement for high-accuracy Power Distribution Network (PDN) and EMI filter verification. Execution is limited to local processes with no network communication detected.
[DYNAMIC_EXECUTION]: The tool dynamically generates SPICE netlist files at runtime, incorporating component values, ESR/ESL parasitics, and PCB trace R+L models. These netlists are used for automated frequency-domain analysis.
[INDIRECT_PROMPT_INJECTION]: The skill ingests schematic and PCB analysis data in JSON format. This represents an indirect prompt injection surface as the agent will consume the resulting risk report; however, the tool operates purely on geometric and numeric data, and no malicious exploitation patterns were observed.
[SAFE]: All external URL references target trusted regulatory and educational domains, including Cornell Law (FCC regulations), IEC standards, and established industry resources like LearnEMC. These are documented for reference and do not involve runtime downloads.

emc