Your next production run will use every footprint in your library.
Has anyone checked them against the datasheets?
On five hand-built prototypes, a wrong land pattern costs you an afternoon and a hot air station. On 500 assembled boards, it costs you the build, the schedule, and the conversation with your CM.
The Pre-Production Library Audit checks your KiCad symbols and footprints against manufacturer datasheets and tells you exactly what will not assemble — before you send the files.

Hi, I am Abbas
I audit KiCad component libraries against manufacturer datasheets.
I got into this the unusual way. I started pulling public KiCad libraries off GitHub and checking their footprints against the datasheets they claimed to implement — partly out of curiosity, partly because I suspected the problem was worse than anyone talks about.
It is. Across 30 libraries and 600 parts, I found 35 land patterns that deviate materially from the manufacturer’s recommendation, and 20 that would cause assembly problems on a real production run. One specific example: part, package, dimension, deviation.
Most of these libraries belong to real companies shipping real boards. Nobody had checked. That’s why this service exists — and it’s why I know what I’m going to find when I open yours.
Is your library actually a problem?
Here’s a question worth asking your team today:
“Who checked the last footprint we added, and against what?”
If the answer is a shrug, or “it worked on the last board,” keep reading.
This is for you if you recognize any of this
- Your library has parts from four different origins — SnapEDA downloads, an easyeda2kicad conversion, a few someone drew at 1am before a deadline, and a handful inherited from somebody’s previous job
- There are two or three symbols for the same MPN and nobody’s certain which one is current
- Two or more engineers commit to the same libraries, and there’s no written rule for what has to be true before a part goes in
- “Verified” in practice means “it assembled fine last time”
- Half your parts have no 3D model, or one that’s rotated 90°
- Nobody’s job title contains the word librarian
The moment this matters most
Four to six weeks before your first assembled production run.
On five hand-built prototypes you can rework a bad footprint with hot air and lose an afternoon. On 500 assembled units you lose the build, the schedule, and six weeks of runway. Everything in your library is about to get manufactured at quantity for the first time — that’s the window where an audit is worth the most and costs the least.
The second-best time is right after a part has already cost you a respin.
This is not for you if
- You’re a one-person team. Your library is consistent because one brain made it. You have real problems; this isn’t one of them.
- You already have a CAD librarian and a written library standard. You’ve solved this. I’d be grading someone’s homework.
- You’re on Altium. Managed component libraries are a feature you’re already paying for. Use it.
- You’re a hobby project or a student. I’d like to help, but this is priced for companies with production budgets.
- You can’t share the library. No NDA gets around ITAR or a hard security policy. Nothing I can do from here.
I’d rather tell you that now than take your money and confirm it later.
What I actually check
Every check below runs against the manufacturer’s datasheet, cited by page and revision. Not against another library. Not against a generic package template.
Land pattern geometry
Pad length, width, pitch, and span compared to the manufacturer’s recommended land pattern. Where the datasheet gives no recommendation, I calculate against IPC-7351B and tell you which density level I used — Least, Nominal, or Most — and why that’s the right one for your assembly process.
Failure mode this catches: the board comes back with opens, bridges, or joints that pass inspection and fail in the field.
Thermal pads and paste apertures
On QFN, DFN, and any part with an exposed pad: paste coverage ratio and aperture pattern. A thermal pad pasted at 100% coverage will float the part and void the joint. This is the single most common serious defect I find.
Failure mode: voiding, tombstoning, parts sitting crooked, thermal performance nowhere near the datasheet.
Solder mask and courtyard
Mask expansion and NSMD/SMD pad definition. Courtyard adequacy at the density level your assembler actually needs. Silkscreen over pad — a DFM violation your fab will either flag or silently strip.
Failure mode: your CM sends the package back and you lose a week, or they fix it silently and you never learn what changed.
Symbol-to-footprint linkage
The right package variant of the right MPN. The correct part number with the wrong package suffix is the mistake that hurts most, because everything downstream — BOM, quote, pick-and-place — is correct and the board is still unbuildable.
Failure mode: 500 correct parts arrive and none of them fit.
Pin mapping and electrical types
Pin numbers and names against the datasheet pinout table. Electrical types — power, input, output, passive, bidirectional — set correctly, so your ERC is actually testing something instead of passing everything.
Failure mode: silent ERC. You think you have a net check. You don’t.
Pin 1 and orientation
Pin 1 marker present, correct, and consistent between symbol, footprint, 3D model, and the orientation your CM expects in the CPL.
Failure mode: an entire reel populated 180° out.
3D models
Present, correctly oriented, correctly offset, dimensionally matching the package drawing.
Failure mode: your mechanical review passes on a model that doesn’t match what actually gets soldered down.
Part identity and supply
MPN validity. Lifecycle status — active, NRND, obsolete. Stock and lead time on anything you’re about to build with.
Failure mode: the footprint is perfect and the part went end-of-life eight months ago.
Library hygiene
Duplicate parts under different names, broken footprint links, orphaned symbols, empty datasheet fields, KLC compliance, naming consistency.
Failure mode: not fatal, but it’s how the other eight problems get in.
What I don’t check
Schematic correctness. Thermal design. Stackup, impedance, or signal integrity. Routing and placement. Firmware. Whether the part is the right choice for your circuit.
I check that the part in your library matches the part in the datasheet. That’s a narrow job, and it’s the whole job.
Who this practice serves
1. Post-lawsuit defendants
If you’ve been named in an ADA Title III website accessibility lawsuit or received a demand letter, I produce the audit documentation your defense attorney needs for settlement and the technical remediation plan your developer needs to meet the conformance obligations in any settlement agreement.
2. Proactive merchants
Over 5,000 ADA digital accessibility lawsuits were filed in the United States in 2025, the majority targeting Shopify and WordPress businesses under $25 million in revenue. If you operate in a commonly-targeted vertical — e-commerce, healthcare, legal services, hospitality — getting ahead of exposure is dramatically cheaper than responding to it after a filing.
3. Web development agencies
If you build Shopify or WordPress sites and don’t offer accessibility services in-house, I work with agencies as their go-to accessibility partner. Your clients get qualified remediation. You get a trusted subcontractor plus a referral arrangement.
Delivery guarantee
You receive your audit report within 10 business days of project kickoff, or the audit is free. No fine print.
I do not make claims about legal outcomes. No vendor can guarantee you won’t be sued again, and any vendor who says otherwise is misrepresenting what’s possible — the Federal Trade Commission fined a major accessibility vendor $1 million in 2025 for exactly that kind of marketing. What I can do is produce documentation that strengthens your defense posture and a remediation process that meaningfully reduces copycat risk.
How I work
Testing methodology
Every audit combines automated scanning (axe DevTools and WAVE), manual keyboard navigation testing, screen reader verification with NVDA, color contrast analysis, and manual code review against WCAG 2.1 Level AA success criteria. Automated tools alone typically catch 30 to 40 percent of real violations — the rest require manual testing, and every audit I deliver includes full manual coverage.
Reporting standard
Deliverables are structured to support defense counsel’s use in settlement negotiations and developers’ use in remediation. Each finding cites the specific WCAG success criterion violated, the page and element where it occurs, the severity level, and concrete remediation guidance.
Frequently Asked Questions
Contact me if you have more questions
Start with a 15-minute call
The fastest way to find out whether I can help is a short conversation. No obligation, no pitch deck — just a direct read on your situation and an honest answer on whether my services are the right fit.
Or direct email: [email protected]
