// Yes, it uses an LLM. No, it’s not a chat window you paste code into — it works inside your live TIA project.
The AI that writes real ladder logic for TIA Portal — and guarantees it compiles.
It reads your open project, writes blocks, imports them, and compiles. If it doesn’t build, the change isn’t applied.
The 70-second tour
Four chapters, one real project: it reads the logic, writes three networks that compile, proves a run against the simulator, and snapshots every change on the way.
Sound familiar?
"Cloning the same block for 12 conveyors by hand, renaming every tag."
→ block cloning"Inherited a 15-year-old project full of STL nobody on the team can read anymore."
→ legacy STL → SCL"Reverse-engineering a block someone else wrote, network by network."
→ code analysis"Translating comments and tags into three languages for the client's HMI."
→ translation"Editing WinCC Unified scripts in a window with zero IntelliSense."
→ WinCC editor"Copy-pasting XML into ChatGPT and back — it never actually sees your project, let alone touches it."
→ works on the live projectWhat Coilmind does
Capability is the proof, not the pitch — every one of these runs against your real, open project.
Reads the project, not just the block
Ask what an FB does, who writes to a tag, what's inconsistent — answered against your real project, not generic PLC trivia. Drop in a screenshot and it reads that too. Beyond chat there are the project dashboards: call graph, architecture and health, instance usage, and which blocks are actually reusable.
Clone a block, rename every tag, done
“Clone FB Motor1 for the second drive, use M2_* tags.” It creates, imports, and compiles — iterating until it builds.
Twelve conveyors in one pass, no model involved
Queue a template against a list — it creates every block, allocates the tags, walks the start addresses and hands you a report of what it made. Deterministic, repeatable, and it keeps a history of every run. This is the fastest answer to the twelfth conveyor, and there is no model anywhere in it.
It does the work — you don't babysit it
You ask in chat; Coilmind writes, imports, compiles, and fixes its own errors. It steps in when the next move is obvious. Every block and screen it touches is snapshotted first, so a change you don’t like is one click back — and every action it takes is listed in the project changelog.
Turn unreadable STL into SCL or LAD
Legacy blocks nobody can maintain, rewritten in chat. You can prove the rewrite behaves like the original by running the original's tests against it — that part needs PLCSIM Advanced. What can't translate cleanly, you see — no silent guesses.
Works with no internet at all
Run a local model right on your machine — on site, in the field, off the network entirely. Your code never leaves the PC. You can work up to 14 days without a connection before the licence needs a short refresh. Local models are weaker than the cloud; how capable is down to your GPU.
Translation that keeps the terminology
Inherited a German or Chinese project? It translates titles, comments, and tag descriptions in the context of the logic — not machine-pidgin.
Reusable block templates — no AI involved
Mark the replaceable tags and parameters; type the new names. Fully deterministic. Works even where AI is off-limits.
Builds HMI screens, not just scripts
Creates screens and screen items, sets dynamizations, and manages HMI tags, tables and groups — then compiles. Scripts get a real editor: Monaco with IntelliSense from the Siemens V21 definitions and project-wide search. It does not create HMI devices, and faceplate types are out of reach for anyone — Openness cannot make them.
Tests that live with the project
Declarative .test.json: set inputs, run N cycles, assert outputs, watch the run live. Coilmind can scaffold the suite from a spec and download to the simulator — never to real hardware. Needs PLCSIM Advanced, which Siemens sells separately.
One session. Nothing cut.
Dry-run protection written into a live pump block, then proved on PLCSIM until it goes green — one session, 9 minutes 36 seconds of recording, nothing cut. The only break is 36 minutes where PLCSIM Advanced was started, and it is labelled in the frame. Typing and generation waits are sped up; nothing else is touched.
If you’ve seen AI demos that fall apart the moment you try them yourself, this is the one to watch.
Your machine, your budget, your TIA version.
Same AI as the big cloud tools — minus their assumptions about your network, your wallet, and your TIA version.
Your code never leaves your machine
Run a local model right on the PC — on site, in the field, even air-gapped. The block templater needs no model at all. Local models are weaker than the cloud, but that ceiling is your GPU budget, not us.
OFFLINE · LOCAL MODELPriced per engineer, month to month
On a card, month to month. No annual lock-in, no org account, no shared seat — one price per engineer. AI is included; no tokens to buy.
PER-ENGINEER · AI INCLUDEDWorks with your existing TIA — V17 to V21
Already on V17 or V18? Coilmind attaches to it — no forced upgrade to the latest portal just to get an assistant. Day-to-day testing runs on V19–V21, and a few conveniences (block renaming among them) need V21, so on the older portals expect the core to work and the edges to be thinner.
V17 → V21Three steps — no complex setup
Download, install, and sign in on your engineering PC. Coilmind sets up Openness access itself — one Windows prompt, no manual whitelisting.
Open your project in TIA Portal, then attach Coilmind to it. Openness can only attach to a project that is already open, so TIA stays in charge.
Chat: analyze, generate, clone, translate, test.
If it gets something wrong, you roll back in one click.
You run production lines. A wrong change is downtime — so autonomy only earns its place next to a safety net. Here’s the mechanics, not a promise.
Built for the working engineer — solo or a small team.
Whether you’re freelancing across clients or part of a 3-person integrator, you do this by hand every week: cloning blocks, reverse-engineering someone else’s logic, translating for the client’s HMI, fixing legacy on site under time pressure.
Take on more projects single-handed — Coilmind does the repetitive build so you ship faster.
Everyone works to the same standard — shared templates, consistent code, no key-person bottleneck.
Built by one automation engineer, not a startup
I spent years writing PLC programs for Siemens hardware, and years doing the same things by hand: cloning blocks for the twelfth conveyor, reverse-engineering someone else’s STL, retyping tag names. Coilmind started as the tool I built for myself.
I’m not a funded startup looking for an exit. svghmi.pro has been selling tools to Siemens engineers — and it runs on subscriptions, not funding rounds. That means Coilmind gets built in the direction users ask for, because there is nobody else to answer to.
If something breaks, you email me and I answer. That’s the whole support org.
support@svghmi.pro
Simple per-licence pricing
Every plan is the full product — AI included, no tokens to buy. Pick the volume that fits; upgrade, downgrade or cancel yourself anytime from your account.
A task is one thing you ask for: "clone this FB for the second drive", "rewrite this STL block in SCL". Coilmind may make a dozen internal calls to finish it — you’re billed for the task, not the calls. Fair use is metered over a rolling 7-day window, so the monthly figure is an average rather than a burst allowance.
Questions you might have
Windows only — the Openness API is Windows-bound. TIA Portal V17 through V21, including the older versions newer cloud assistants don’t support. Day-to-day testing runs on V19–V21 and a few conveniences need V21, so on V17 and V18 expect the core to work and the edges to be thinner.
One condition worth knowing before you download: Openness can only attach to a project that is already open in TIA Portal. Coilmind cannot open a project for you — you open it, then attach.