PartReel — open KiCad parts registry for humans & AI agents

MCP serverAI & models

Partreel gives your AI access to a library of verified KiCad footprints, symbols, and 3D models for electronic components. Once added, your AI can look up quality-checked part files for circuit board designs without any signup. Files are community curated and free to reuse under the CC-BY-4.0 license.

Use PartReel — open KiCad parts registry for humans & AI agents in Claude, ChatGPT or Ahel Desktop

Free. Sign in, add PartReel — open KiCad parts registry for humans & AI agents and connect your AI. About a minute.

Also: Claude Code · Cursor · Codex

Then ask your AI: use PartReel to search parts

Details

Available today. Use it from your connected AI after setup.

Add Partreel, then ask your AI for the footprint, symbol, or 3D model of a component you are working with. It can also pull files for a whole list of parts from your board design.

PartReel — open KiCad parts registry for humans & AI agentsStart free

What your AI can do with it

  • Find verified footprints for electronic components
  • Look up schematic symbols for specific parts
  • Retrieve 3D models of components
  • Fetch part files without creating an account
  • Reuse downloaded files under the CC-BY-4.0 license

Install PartReel — open KiCad parts registry for humans & AI agents

The server’s own address, for the clients that take one directly. Or connect Ahel once and every client you use reads it from one address, with the account kept on Ahel rather than in each client’s config.

  • Claude Code

    claude mcp add --transport http --scope user partreel-open-kicad-parts-regist 'https://mcp.partreel.com/mcp'

    Run it once in your project, then open /mcp to approve any sign-in the server asks for.

  • Claude Desktop

    https://mcp.partreel.com/mcp

    Add a custom connector in Settings, paste this address, and approve the sign-in.

  • Cursor

    cursor://anysphere.cursor-deeplink/mcp/install?name=partreel-open-kicad-parts-regist&config=eyJ1cmwiOiJodHRwczovL21jcC5wYXJ0cmVlbC5jb20vbWNwIn0=

    Open the link and Cursor adds the server at that address.

  • ChatGPT

    https://mcp.partreel.com/mcp

    In Settings, enable Developer mode, create an MCP app, and paste this address. Your plan and workspace must allow custom apps.

  • Codex

    codex mcp add partreel-open-kicad-parts-regist --url 'https://mcp.partreel.com/mcp'

    Run it once, then sign in with codex mcp login partreel-open-kicad-parts-regist if the server asks for an account.

From the project's README

As published by mingyo186/partreel in README.md.

No login. No paywall. Machine-verified. Every part ships as a bundle of KiCad footprint (.kicad_mod) + symbol (.kicad_sym) + 3D model (STEP + GLB), generated from manufacturer-datasheet dimensions and checked by automated quality gates on every deploy.

Site: https://partreel.com · API: https://partreel.com/api/ · MCP server: https://mcp.partreel.com/mcp · Agent guide: https://partreel.com/agents/

Why this exists

  • The official KiCad library is excellent — use it when it has your part. PartReel focuses on parts the official library doesn't cover (checked against GitLab master before we add anything).
  • Aggregator sites lock downloads behind sign-ups; agents and scripts can't use them. Everything here is a plain URL.
  • AI agents regenerate footprints from scratch every time — wasteful and unverified. PartReel is the place to fetch a verified part instead, and report back (field_reports in the API).

Provenance — verify us, don't trust us

Every part's API entry (/api/v1/parts/<id>.json) carries machine-readable provenance: the exact datasheet figure/table each dimension came from (dimensions_source), the generator source file in this repo, the quality gates it passed, and field-usage reports. Gates run in CI on every push and block deployment on failure — see .github/workflows/deploy.yml.

GateChecks
validate_kicad.pys-expression structure, pads, layers, pin-1
KLC drawing rulessilk 0.12 / fab 0.10 + pin-1 chamfer / courtyard 0.05
check_overlap.pyno overlapping text in previews
check_render.pyfiles exist, SVG pads/outlines match source, pages/API present, datasheet cited
check_zfight.pyno coplanar z-fighting faces, pins are individual bodies
validate_step.pySTEP solids pass FreeCAD kernel isValid

For AI agents

  • Static JSON API, no auth, no rate limit: /api/v1/parts.json
  • Remote MCP server (nothing to install): claude mcp add --transport http partreel https://mcp.partreel.com/mcp — tools: search_parts, get_part, list_parts, submit_part, report_feedback, request_part, how_to_contribute
  • Machine guide: /llms.txt
  • The loop: search → use → if a part is missing, build it from the datasheet and submit_part — it is shared with every agent instantly (labeled unverified/staging), then hourly CI gates + review promote it to verified.
  • request_part generates parametric families (pin headers, LDOs) server-side in ~5 min; the PR path (CONTRIBUTING-AGENTS.md) remains for 3D models and edits.

Repository layout

library/                 part assets (CC-BY-4.0): .kicad_mod / .kicad_sym / .step / .glb / meta.json
generators/              parametric generators (Python; KiCad text + FreeCAD 3D) + quality gates
api/, p/, assets/        static site & JSON API (built by CI)
mcp/                     Cloudflare Worker MCP server

License

  • Code (generators, site): MIT
  • Part assets (library/): CC-BY-4.0

Dimensions come from manufacturer datasheets and are provided as-is; verify before manufacturing.

Credits

Bug reports are contributions. Reporters whose feedback led to a fix are recorded in CREDITS.md and in the fix commits as Reported-by:; imported-library authors are credited in ATTRIBUTIONS.md.

Parts index

Currently 18258 parts, all machine-verified (structure, KLC drawing rules, render completeness, 3D coplanar/merged-pin checks, STEP kernel) with datasheet-cited dimensions.

connector (4330)

PartMPNManufacturerFiles
004500 (CERN library)004500ERNIcern_erni_004500
02 01 160 2101 (CERN library)02 01 160 2101HARTINGcern_harting_02_01_160_2101
02 01 160 2102 (CERN library)02 01 160 2102HARTINGcern_harting_02_01_160_2102
02 02 160 2201 (CERN library)02 02 160 2201HARTINGcern_harting_02_02_160_2201
02 02 160 2301 (CERN library)02 02 160 2301HARTINGcern_harting_02_02_160_2301
02 04 160 1101 (CERN library)02 04 160 1101HARTINGcern_harting_02_04_160_1101
02 07 160 1101 (CERN library)02 07 160 1101HARTINGcern_harting_02_07_160_1101
031-6575 (CERN library)031-6575AMPHENOL-RFcern_amphenol_031_6575
054392 (CERN library)054392ERNIcern_erni_054392
054546 (CERN library)054546ERNIcern_erni_054546
054594 (CERN library)054594ERNIcern_erni_054594
054595 (CERN library)054595ERNIcern_erni_054595
054596 (CERN library)054596ERNIcern_erni_054596
063197 (CERN library)063197ERNIcern_erni_063197
063209 (CERN library)063209ERNIcern_erni_063209
063210 (CERN library)063210ERNIcern_erni_063210
064317 (CERN library)064317ERNIcern_erni_064317
064318 (CERN library)064318ERNIcern_erni_064318
064321 (CERN library)064321ERNIcern_erni_064321
09 02 164 2921 (CERN library)09 02 164 2921HARTINGcern_harting_09_02_164_2921
09 02 264 6825 (CERN library)09 02 264 6825HARTINGcern_harting_09_02_264_6825
09 02 264 6850 (CERN library)09 02 264 6850HARTINGcern_harting_09_02_264_6850
09 03 000 6104 (CERN library)09 03 000 6104HARTINGcern_harting_09_03_000_6104
09 03 000 6110 (CERN library)09 03 000 6110HARTINGcern_harting_09_03_000_6110
09 03 000 6162 (CERN library)09 03 000 6162HARTINGcern_harting_09_03_000_6162
09 03 000 6182 (CERN library)09 03 000 6182HARTINGcern_harting_09_03_000_6182
09 03 000 6225 (CERN library)09 03 000 6225HARTINGcern_harting_09_03_000_6225
09 03 000 6262 (CERN library)09 03 000 6262HARTINGcern_harting_09_03_000_6262
09 03 000 6265 (CERN library)09 03 000 6265HARTINGcern_harting_09_03_000_6265
09 03 124 6901 (CERN library)09 03 124 6901HARTINGcern_harting_09_03_124_6901
09 03 142 6901 (CERN library)09 03 142 6901HARTINGcern_harting_09_03_142_6901
09 03 160 6901 (CERN library)09 03 160 6901HARTINGcern_harting_09_03_160_6901
09 03 164 2921 (CERN library)09 03 164 2921HARTINGcern_harting_09_03_164_2921
09 03 164 6921 (CERN library)09 03 164 6921HARTINGcern_harting_09_03_164_6921
09 03 164 6922 (CERN library)09 03 164 6922HARTINGcern_harting_09_03_164_6922
09 03 164 6951 (CERN library)09 03 164 6951HARTINGcern_harting_09_03_164_6951
09 03 178 7901 (CERN library)09 03 178 7901HARTINGcern_harting_09_03_178_7901
09 03 196 2921 (CERN library)09 03 196 2921HARTINGcern_harting_09_03_196_2921
09 03 196 6921 (CERN library)09 03 196 6921HARTINGcern_harting_09_03_196_6921
09 03 224 6805 (CERN library)09 03 224 6805HARTINGcern_harting_09_03_224_6805
09 03 224 6830 (CERN library)09 03 224 6830HARTINGcern_harting_09_03_224_6830
09 03 224 6864 (CERN library)09 03 224 6864HARTINGcern_harting_09_03_224_6864
09 03 224 6864 (CERN library)09 03 224 6864HARTINGcern_harting_09_03_224_6864_a
09 03 232 6825 (CERN library)09 03 232 6825HARTINGcern_harting_09_03_232_6825
09 03 232 6850 (CERN library)09 03 232 6850HARTINGcern_harting_09_03_232_6850
09 03 242 6805 (CERN library)09 03 242 6805HARTINGcern_harting_09_03_242_6805
09 03 242 6805 (CERN library)09 03 242 6805HARTINGcern_harting_09_03_242_6805_a
09 03 260 6804 (CERN library)09 03 260 6804HARTINGcern_harting_09_03_260_6804
09 03 264 6825 (CERN library)09 03 264 6825HARTINGcern_harting_09_03_264_6825
09 03 264 6850 (CERN library)09 03 264 6850HARTINGcern_harting_09_03_264_6850
09 03 278 6804 (CERN library)09 03 278 6804HARTINGcern_harting_09_03_278_6804
09 03 296 2825 (CERN library)09 03 296 2825HARTINGcern_harting_09_03_296_2825
09 03 296 2850 (CERN library)09 03 296 2850HARTINGcern_harting_09_03_296_2850
09 03 296 6824 (CERN library)09 03 296 6824HARTINGcern_harting_09_03_296_6824
09 03 296 6825 (CERN library)09 03 296 6825HARTINGcern_harting_09_03_296_6825
09 03 296 6850 (CERN library)09 03 296 6850HARTINGcern_harting_09_03_296_6850
09 03 296 6861 (CERN library)09 03 296 6861HARTINGcern_harting_09_03_296_6861
09 03 478 6864 (CERN library)09 03 478 6864HARTINGcern_harting_09_03_478_6864
09 03 696 6921 (CERN library)09 03 696 6921HARTINGcern_harting_09_03_696_6921
09 03 796 6825 (CERN library)09 03 796 6825HARTINGcern_harting_09_03_796_6825
09 03 796 6850 (CERN library)09 03 796 6850HARTINGcern_harting_09_03_796_6850
09 04 132 2921 (CERN library)09 04 132 2921HARTINGcern_harting_09_04_132_2921
09 04 132 6921 (CERN library)09 04 132 6921HARTINGcern_harting_09_04_132_6921
09 04 232 6826 (CERN library)09 04 232 6826HARTINGcern_harting_09_04_232_6826
09 04 232 6831 (CERN library)09 04 232 6831HARTINGcern_harting_09_04_232_6831
09 05 148 2931 (CERN library)09 05 148 2931HARTINGcern_harting_09_05_148_2931
09 05 248 6831 (CERN library)09 05 248 6831HARTINGcern_harting_09_05_248_6831
09 06 031 2921 (CERN library)09 06 031 2921HARTINGcern_harting_09_06_031_2921
09 06 048 2905 (CERN library)09 06 048 2905HARTINGcern_harting_09_06_048_2905
09 06 115 2911 (CERN library)09 06 115 2911HARTINGcern_harting_09_06_115_2911
09 06 131 6922 (CERN library)09 06 131 6922HARTINGcern_harting_09_06_131_6922
09 06 131 6924 (CERN library)09 06 131 6924HARTINGcern_harting_09_06_131_6924
09 06 132 6931 (CERN library)09 06 132 6931HARTINGcern_harting_09_06_132_6931
09 06 148 2901 222 (CERN library)09 06 148 2901 222HARTINGcern_harting_09_06_148_2901_222
09 06 148 6901 (CERN library)09 06 148 6901HARTINGcern_harting_09_06_148_6901
09 06 203 2811 (CERN library)09 06 203 2811HARTINGcern_harting_09_06_203_2811
09 06 215 2811 (CERN library)09 06 215 2811HARTINGcern_harting_09_06_215_2811
09 06 215 2821 (CERN library)09 06 215 2821HARTINGcern_harting_09_06_215_2821
09 06 215 2854 (CERN library)09 06 215 2854HARTINGcern_harting_09_06_215_2854
09 06 215 2856 (CERN library)09 06 215 2856HARTINGcern_harting_09_06_215_2856
09 06 231 2881 (CERN library)09 06 231 2881HARTINGcern_harting_09_06_231_2881
09 06 231 6822 (CERN library)09 06 231 6822HARTINGcern_harting_09_06_231_6822
09 06 231 6822 (CERN library)09 06 231 6822HARTINGcern_harting_09_06_231_6822_a
09 06 248 6826 (CERN library)09 06 248 6826HARTINGcern_harting_09_06_248_6826
09 06 248 6832 (CERN library)09 06 248 6832HARTINGcern_harting_09_06_248_6832
09 06 248 6834 (CERN library)09 06 248 6834HARTINGcern_harting_09_06_248_6834
09 06 248 6836 (CERN library)09 06 248 6836HARTINGcern_harting_09_06_248_6836
09 06 248 7833 (CERN library)09 06 248 7833HARTINGcern_harting_09_06_248_7833
09 12 000 9905 (CERN library)09 12 000 9905HARTINGcern_harting_09_12_000_9905
09 12 000 9905 (CERN library)09 12 000 9905HARTINGcern_harting_09_12_000_9905_alt
09 12 000 9908 (CERN library)09 12 000 9908HARTINGcern_harting_09_12_000_9908
09 12 012 9901 (CERN library)09 12 012 9901HARTINGcern_harting_09_12_012_9901
09 16 000 9908 (CERN library)09 16 000 9908HARTINGcern_harting_09_16_000_9908
09 16 000 9918 (CERN library)09 16 000 9918HARTINGcern_harting_09_16_000_9918
09 16 000 9918 (CERN library)09 16 000 9918HARTINGcern_harting_09_16_000_9918_alt
09 18 516 6324 (CERN library)09 18 516 6324HARTINGcern_harting_09_18_516_6324
09 18 516 6324 (CERN library)09 18 516 6324HARTINGcern_harting_09_18_516_6324_alt
09 19 564 6824 (CERN library)09 19 564 6824HARTINGcern_harting_09_19_564_6824
09 19 564 6824 (CERN library)09 19 564 6824HARTINGcern_harting_09_19_564_6824_alt
09 22 116 6921 (CERN library)09 22 116 6921HARTINGcern_harting_09_22_116_6921
09 22 132 6921 (CERN library)09 22 132 6921HARTINGcern_harting_09_22_132_6921
09 22 232 6825 (CERN library)09 22 232 6825HARTINGcern_harting_09_22_232_6825
09 22 232 6850 (CERN library)09 22 232 6850HARTING[cern_h

Shortened here. Read the whole README on GitHub.

Tools it offers (7)

What this server listed when Ahel dialed its public endpoint in Oct 2026, with no key and no account of yours. The names are the server’s own.

  • search_parts
  • get_part
  • list_parts
  • report_feedback
  • request_part
  • how_to_contribute
  • submit_part

Signals

GitHub stars
2
Forks
1
Last commit
Sep 2026
Advanced
Delivery
partreel MCP server → your Ahel connector (mcp.ahel.ai) → your AI.
Item type
mcp-server
Key
io-github-mingyo186-partreel
Source
github.com/mingyo186/partreel
Hosted endpoint
https://mcp.partreel.com/mcp