Your AI Agent Can Now Read Schematics and Catch PCB Bugs
It gives AI coding agents the domain knowledge to read KiCad schematics, audit PCB layouts for EMC and signal integrity issues, and catch costly hardware bugs before fabrication.

What it does
kicad-happy is a set of AI agent skills that lets Claude Code, OpenAI Codex, and similar tools understand KiCad electronics projects at a deep level. It parses schematics, PCB layouts, and Gerber files directly into structured data, then traces power trees, detects subcircuits, validates passive networks, and flags design errors. The skills also cover the full prototype-to-production workflow: SPICE simulation, EMC pre-compliance checks, datasheet extraction, BOM sourcing from major distributors, and fabrication prep.
The interesting bit
The parser is pure Python with zero dependencies and does not require KiCad itself to be installed; it reads raw S-expression files and reconstructs design intent well enough to compute feedback divider ratios and spot missing ESD protection. It essentially turns a general-purpose coding agent into a domain-specific hardware reviewer that can audit connectors for grounding issues or suggest whether your board needs FCC Part 15 certification.
Key highlights
- Works as plugins for Claude Code, OpenAI Codex, GitHub Copilot CLI, Google Antigravity, and opencode, or as a GitHub Action for automated PR reviews
- Parses schematics and PCBs into structured netlists, builds power trees, identifies functional subcircuits, and validates passive component networks such as RC filters, crystal load caps, and divider ratios
- Runs 44 EMC pre-compliance rule checks covering radiated emissions, PDN impedance, differential pair skew, and ESD paths
- Generates SPICE testbenches for detected subcircuits and runs Monte Carlo tolerance analysis via ngspice, LTspice, or Xyce
- Manages BOM lifecycle, searches DigiKey/Mouser/LCSC/JLCPCB, extracts structured specs from datasheet PDFs, and checks production readiness
Caveats
- Claude Code’s
/plugin updatemay fail to detect new versions due to a known upstream issue, requiring manual cache clearing - Release candidates have passed regression tests but explicitly have not completed extended manual validation
- Windows manual installs require Developer Mode or elevated privileges to create the necessary symbolic links
Verdict
Hardware designers already using KiCad and AI coding agents should grab this; it effectively adds a first-pass design review and procurement assistant to your workflow. If you don’t work with electronics or don’t use agentic coding tools, this is a very specific set of plugins to skip.
Frequently asked
- What is aklofas/kicad-happy?
- It gives AI coding agents the domain knowledge to read KiCad schematics, audit PCB layouts for EMC and signal integrity issues, and catch costly hardware bugs before fabrication.
- Is kicad-happy open source?
- Yes — aklofas/kicad-happy is open source, released under the MIT license.
- What language is kicad-happy written in?
- aklofas/kicad-happy is primarily written in Python.
- How popular is kicad-happy?
- aklofas/kicad-happy has 1.2k stars on GitHub.
- Where can I find kicad-happy?
- aklofas/kicad-happy is on GitHub at https://github.com/aklofas/kicad-happy.