topic Can I run Scripts & Add-Ons from Electronic Design? in Fusion Electronics Forum

Can I run Scripts & Add-Ons from Electronic Design?

jasonchildress1978 — Tue, 15 Apr 2025 12:46:22 GMT

I'd like to start automating some of my electronic design processes with python. But I'm having a hard time finding a place to start.

I turned on the Scripts and Add-Ins feature in preferences, but it appears that I can only use it in a regular 3D design and not my electronic design? The dialog can only be opened from the Utilities Tab (in a regular design) and its modal so once I open, I can't switch to my electronic design.

Is there currently a way to run python scripts in an electronic design to help automate my workflow?

Re: Can I run Scripts & Add-Ons from Electronic Design?

jorge_garcia — Tue, 15 Apr 2025 21:57:51 GMT

Hello @jasonchildress1978 ,

I hope you're doing well. Currently, Electronics is not properly supported in Fusion Python API. There is some functionality available which is what the Ultra-librarian and SnapEDA plugins use, but there isn't much official documentation for using the Fusion Python API in Electronics. For now the recommended approach is to use the ULP language and scripts.

Fusion Electronics ships with several of these as examples. You'll find the documentation here:

https://help.autodesk.com/view/fusion360/ENU/?guid=ECD-AUTOMATION-CPT

Let me know if there's anything else I can do for you.

Best Regards,

Re: Can I run Scripts & Add-Ons from Electronic Design?

jasonchildress1978 — Mon, 21 Apr 2025 11:44:33 GMT

Thanks. I'll look into using a ULP script.

Are there future plans to support Python in electronics designs?

Re: Can I run Scripts & Add-Ons from Electronic Design?

jorge_garcia — Mon, 21 Apr 2025 17:26:52 GMT

Hello @jasonchildress1978 ,

I hope you're doing well. I have added your comments to an existing ticket for getting Electronics into the Python API, it something we definitely want to do but it just hasn't been prioritized yet.

Let me know if there's anything else I can do for you.

Best Regards,

Re: Can I run Scripts & Add-Ons from Electronic Design?

holla2040 — Sat, 07 Jun 2025 05:40:44 GMT

I'd like to see if projects like these

https://github.com/ArchimedesCrypto/fusion360-mcp-server

https://github.com/Joelalbon/Fusion-MCP-Server

could connect to the future electronics API via MCP.

Then have an AI, like claude, pull the schematic parts, nodes and nets then crunch on optimizing brd parts placement with these criteria - minimizing board space, minimizing signal length, analog/digital separation etc. I can see this happening once an electronics API exists and someone writes an MCP server for it.

https://www.anthropic.com/news/model-context-protocol

Re: Can I run Scripts & Add-Ons from Electronic Design?

pal.szabo — Tue, 10 Jun 2025 09:41:49 GMT

Fully agree. We would actually do half the development for you guys, in the form of add-ins with an API access. I would love to see this prioritized.

Best regards,

Paul

Re: Can I run Scripts & Add-Ons from Electronic Design?

josh_l — Tue, 10 Mar 2026 10:45:43 GMT

Agree... I desperately want to give Claude access to my PCB designs through an API. It will be hard to resist switching to KiCad at some point, given how much time it will save to be able to use Claude Code to review, tweak and even co-author new PCB designs. I should be able to verify a schematic against a natural language document of constraints and design choices.

Re: Can I run Scripts & Add-Ons from Electronic Design?

holla2040 — Tue, 10 Mar 2026 15:25:39 GMT

Agree, I've used Claude Code for multiple hours a day for 10 months now and would just love to automate some of tedium. I will say CC does a great job with ULPs.

Re: Can I run Scripts & Add-Ons from Electronic Design?

holla2040 — Tue, 10 Mar 2026 15:30:09 GMT

My Thomson project is just waiting for the day when fusion electronics has an API.

https://github.com/holla2040/ThomsonLint

Re: Can I run Scripts & Add-Ons from Electronic Design?

jorge_garcia — Tue, 10 Mar 2026 20:25:11 GMT

Hi everyone,

Just a small update, the Electronics API is in active development. It will take some time to complete but it's no longer just a plan, it's being worked on.

Let me know if there's anything else I can do for you.

Best Regards,

Re: Can I run Scripts & Add-Ons from Electronic Design?

holla2040 — Tue, 10 Mar 2026 20:44:39 GMT

I made fantastic progress on my ThomsonLint AI design reviewer today.

https://github.com/holla2040/ThomsonLint

● ThomsonLint — AI-powered hardware design review using structured rules instead of vibes.

You export your schematic + board data from Fusion Electronics as JSON (via a ULP that runs in 2 clicks), then point
Claude Code (or Gemini) at it along with a knowledge base of 69+ hardware design rules covering power, high-speed,
EMC, thermal, and DFM domains.

The AI checks your design against specific rules and cites exactly which ones you're violating — not just "this looks
wrong" but "this violates rule EMC_ESD_001 because your connector has no TVS protection."

What's in the box:
- Ontology of hardware design rules (JSON, machine-readable)
- Knowledge base with explanations and examples
- Fusion Electronics ULP exporter (schematic + board → JSON)
- Works with Claude Code CLI or Gemini web UI

git clone github.com/holla2040/ThomsonLint

cd ThomsonLint

claude

Then ran the fusion-electronics-export.ulp in the sch and brd context.

Here's an example running this command in Claude Code

Read the review instructions in review_instructions.txt, then review my design
using the exported JSON files and datasheets in exports/. Please begin the review.

Here's Claude's output

--- ThomsonLint Design Review: Pendant 2 I/O Board Design Summary A 2-layer industrial I/O board (~190mm x 110mm) with: - Power: 24V input → K7805-1000R3 DC-DC → 5V, external 3.3V source - I/O Expansion: PCF8574 (I2C) driving ULN2003 relay driver → 4x SPDT relays - Comms: MAX3232 (RS-232), RS-485 transceiver, 2x LTV-352T optocouplers, 2x MB10S bridge rectifiers - Connectors: DB9 F/M, USB-C (power only), JST headers, screw terminals, 18650 battery holder --- CRITICAL Issues 1. U5 (MAX3232) GND is NOT connected to board GND Rule: PWR_DECPL_001 | Severity: CRITICAL The MAX3232's GND pin is on net N$35, which connects only to C9.2, C10.2, and C11.2 (the charge pump and VCC decoupling caps). This net is never connected to the main GND net. The charge pump capacitors for V+, V-, and the VCC bypass all have their "ground" side floating relative to the rest of the circuit. This means the MAX3232 has no DC return path to ground. The RS-232 interface will not function. Fix: Connect U5 GND (net N$35) to the main GND net. --- 2. No 3.3V regulator on board — 3V3 rail has no source Rule: PWR_DECPL_001 | Severity: CRITICAL The 3V3 net powers U2 (PCF8574), U5 (MAX3232), U6 (RS-485), and pull-up resistors, but no component on this board generates 3.3V. The only possible sources are external connectors H1 and H3 (pin 1 = 3V3). If the external source is not connected, the entire digital section is unpowered. The bottom copper layer has a 3V3 polygon pour, which will carry this rail — but without a source on-board, the design depends entirely on an external supply. Recommendation: If intentional, add clear documentation. Consider adding a 3.3V LDO from the 5V rail for standalone operation or as a backup. --- 3. U6 (RS-485) !RE and DE permanently tied to 3V3 Rule: SCH_CONN_002 | Severity: CRITICAL U6's !RE (active-low receive enable) and DE (active-high drive enable) are both tied to 3V3. This permanently enables the transmitter and permanently disables the receiver. The RS-485 port can only transmit, never receive. On a multi-drop RS-485 bus, a permanently-enabled driver will also block all other devices from transmitting. Fix: Route DE and !RE to a GPIO (e.g., a spare PCF8574 output or an off-board MCU pin) so the transceiver direction can be controlled. --- MAJOR Issues 4. DB9 connectors (J13, J14) have no ESD protection Rule: EMC_ESD_001 | Severity: Major J13 (DB9-F) and J14 (DB9-M) are external-facing RS-232 connectors with no TVS or ESD protection on any signal lines. RS-232 ports are frequently exposed to ESD events from cable handling. Fix: Add TVS diode arrays (e.g., PRTR5V0U2X or equivalent) on the RS-232 signal lines near the connectors. 5. USB-C connector (J22) has no ESD protection on VBUS Rule: EMC_ESD_001 | Severity: Major J22 USB-C VBUS connects directly to the 5V0 rail with no TVS protection. The SHIELD pin goes to GND. While the USB data pins appear unused (power-only configuration), VBUS still needs protection. Fix: Add a TVS diode on VBUS (e.g., SMBJ5.0A or similar). 6. USB-C VBUS connected directly to 5V0 — potential back-feed Rule: PWR_PROT_001 | Severity: Major J22 VBUS is directly connected to the 5V0 rail, which is also the output of U8 (K7805 regulator). If USB is plugged in while 24V is active, two sources will fight on the 5V rail. If 24V is not present, USB power could back-feed into U8's output pin. Fix: Add a Schottky diode (e.g., BAT54) or ideal diode controller to OR the USB and regulator 5V sources, or add documentation that they are mutually exclusive. 7. Battery (BT1) has no protection circuitry Rule: PWR_PROT_001 | Severity: Major The 2-cell 18650 battery holder has both cells wired in parallel (both + pins on N$42, both - pins on GND). There is: - No reverse polarity protection - No over-discharge protection - No over-current protection (only the connector J3 between battery and the external circuit) - No cell balancing (parallel cells with no balancing resistors) If mismatched cells are inserted, high circulating currents can occur. Fix: Add a battery protection IC or at minimum individual cell fuses. 8. No bulk decoupling on U8 (K7805) input/output with ceramic caps Rule: PWR_DECPL_001 | Severity: Major U8 input has C12 and output has C13 (both 10uF@50V electrolytic). DC-DC converters like the K7805 require ceramic capacitors (typically 10-22uF MLCC) close to input and output pins for high-frequency decoupling. Electrolytics alone have too much ESR and ESL for proper converter operation. Fix: Add 10uF or 22uF ceramic capacitors (0805 or 1206) in parallel with C12 and C13, placed close to U8. 9. Many components missing values/part numbers Rule: SCH_ATTR_001 | Severity: Major Several components have empty or missing value fields: - D2, D3 (TVS): No voltage rating specified - D6, D7, D8, D9 (diodes): No part number or voltage/current rating - Q3 (P-ch FET): No part number specified (only LCSC C393520) - Q1, Q2, Q4, Q5 (depletion N-ch FETs): Only LCSC number, no MPN - Multiple capacitors (C1-C3, C7-C11, C15): No LCSC or MPN, only values This makes BOM generation unreliable and prevents proper design verification. --- MINOR Issues 10. Only 2 fiducial markers (JP1, JP2) Rule: DFM_FID_001 | Severity: Minor Only 2 fiducials are placed (at opposite corners). Industry standard recommends 3 fiducials in an L-pattern for accurate pick-and-place alignment. Fix: Add a third fiducial marker. 11. I2C pull-ups (R8, R9) are marked DNP Rule: SCH_CONN_002 | Severity: Minor The I2C bus (SCL/SDA) pull-up resistors R8 and R9 are both DNP. This means the I2C bus relies on pull-ups being provided elsewhere (likely on the external MCU board connected via H1). If the external board doesn't provide pull-ups, I2C communication will fail. Fix: Populate R8/R9 with 4.7k or 10k, or ensure the external controller board provides pull-ups. 12. Multiple empty/orphan nets Rule: SCH_NET_001 | Severity: Minor Nets N$27, N$62-N$76 have no pins connected. These are leftover artifacts and should be cleaned up to avoid confusion. 13. 24V0 net not flagged as power net Rule: SCH_ATTR_001 | Severity: Minor The 24V0 net (is_power: false) carries the main 24V power rail but is not classified as a power net. This may cause DRC tools to miss power-integrity checks. 14. GND has zero trace segments — relies entirely on copper pour Rule: PWR_DECPL_001 | Severity: Advisory The GND signal has trace_length_mm: 0 and segment_count: 0, meaning it relies 100% on the top-layer ground pour. While copper pours are good for ground, verify that all GND pads have adequate thermal relief connections and that no isolated ground islands exist, especially around the dense connector/relay area. 15. 3V3 trace width only 0.254mm (10mil) for a power rail Rule: PWR_TRACE_001 | Severity: Advisory The 3V3 rail uses 0.254mm traces with 7 vias to transition to the bottom-layer 3V3 pour. While current may be low, power traces should generally be wider. Verify current requirements for U2 + U5 + U6 combined. --- Summary Table ┌─────┬──────────┬───────────────┬──────────────┬────────────────────────────────────────────┐ │ # │ Severity │ Rule │ Component │ Issue │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 1 │ CRITICAL │ PWR_DECPL_001 │ U5 (MAX3232) │ GND pin not connected to board GND │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 2 │ CRITICAL │ PWR_DECPL_001 │ 3V3 rail │ No on-board 3.3V source │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 3 │ CRITICAL │ SCH_CONN_002 │ U6 (RS-485) │ !RE/DE permanently tied — transmit-only │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 4 │ Major │ EMC_ESD_001 │ J13, J14 │ No ESD on DB9 RS-232 connectors │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 5 │ Major │ EMC_ESD_001 │ J22 │ No ESD on USB-C VBUS │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 6 │ Major │ PWR_PROT_001 │ J22/U8 │ USB VBUS back-feed to regulator │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 7 │ Major │ PWR_PROT_001 │ BT1 │ No battery protection circuitry │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 8 │ Major │ PWR_DECPL_001 │ U8 │ No ceramic caps on DC-DC converter │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 9 │ Major │ SCH_ATTR_001 │ Multiple │ Missing component values/part numbers │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 10 │ Minor │ DFM_FID_001 │ JP1/JP2 │ Only 2 fiducials (need 3) │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 11 │ Minor │ SCH_CONN_002 │ R8, R9 │ I2C pull-ups are DNP │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 12 │ Minor │ SCH_NET_001 │ N$27, N$62+ │ Orphan nets with no connections │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 13 │ Minor │ SCH_ATTR_001 │ 24V0 │ Not flagged as power net │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 14 │ Advisory │ PWR_DECPL_001 │ GND │ No traces, pour-only — verify connectivity │ ├─────┼──────────┼───────────────┼──────────────┼────────────────────────────────────────────┤ │ 15 │ Advisory │ PWR_TRACE_001 │ 3V3 │ Narrow traces for power rail │ └─────┴──────────┴───────────────┴──────────────┴────────────────────────────────────────────┘ Issues found: 3 Critical, 6 Major, 4 Minor, 2 Advisory The most urgent fix is Issue #1 — the MAX3232 GND isolation will cause complete RS-232 failure. Issues #2 and #3 should also be addressed before fabrication. ✻ Cooked for 2m 39s

Re: Can I run Scripts & Add-Ons from Electronic Design?

josh_l — Wed, 11 Mar 2026 10:18:47 GMT

This is so cool. I'm going to give it a run when I think my current project is ready to review. It's exciting to think that once the Fusion Electronics API is working you could then just ask Claude to make the changes directly.

We live in the future!

Re: Can I run Scripts & Add-Ons from Electronic Design?

pal.szabo — Mon, 16 Mar 2026 22:16:27 GMT

This is a fantastic addition to any final board checklist before fabrication! Thank you very much for sharing!

Best regards,

Paul

Re: Can I run Scripts & Add-Ons from Electronic Design?

pal.szabo — Mon, 18 May 2026 21:06:33 GMT

I also couldn't wait for the API, so I hacked my way around it using ULP's and scripts. There is an add-in I have been working on for a couple of months now and it just released on the Autodesk App Store. It's a collection of utilities and custom ULPs ported to a UI-friendly interface. It's free, open-source and available on GitHub, if anyone is interested.

Can't wait to finally be able to access Fusion-specific data via an API, though... like the stackup, the rules, the selection filters, the design manager data, etc.

Best regards.

Paul

Re: Can I run Scripts & Add-Ons from Electronic Design?

holla2040 — Fri, 29 May 2026 14:07:05 GMT

Sorry for late reply. This looks great, very helpful. Thanks for the add-in.