The SEC and DOJ are now prosecuting AI-washing. Diligence caught up in 2025 — did your process? →
Buyers deduct 10–20% of enterprise value for AI provenance gaps →

Find it before
the buyer does.

Assay Ready runs the buyer's diligence engine against your own codebase — months before the process starts. You get a remediation punch-list instead of a deal recommendation, so findings get fixed on your timeline instead of priced into their offer.

assay . --mode seller

Assay Ready — Pre-Diligence Punch-List

Findings a buyer's screen would surface: 3 · categories not assessed: 1
1 · HIGHTruck factor is 1 — knowledge concentrated in one personconfidence 80%

What a buyer reads into it: the business is one resignation away from a maintenance cliff, and prices retention risk into the offer.

How to close it: pair-review the single-owned files and land real commits from a second engineer. Evidence that someone else already ships in those files is what moves the number.

GAPDependency vulnerabilities: not assessed — UNKNOWN, not clean. A buyer running a fuller scope may assess these.
10–20%of enterprise value deducted for AI provenance gaps — the discount you can pre-empt
0.5–1.5xmultiple reduction buyers apply for key-person risk
20%of dealmakers walked away over AI concerns in the past year
Why sellers run it

Every finding you close early is a discount you never have to argue about.

In diligence, a finding surfaces at the worst possible moment: after the LOI, when the buyer has leverage and you have a closing timeline. The same finding, closed six months earlier, is just an engineering ticket. Assay Ready moves the discovery to the side of the table where it's cheap.

The same engine, nothing softened

This is not a friendlier scan. It's the identical analyzer battery a buyer runs, with the identical thresholds — because a report that flatters you is worthless the moment their engine disagrees with it.

Ordered by negotiating cost

The punch-list is sequenced by what a finding costs you in a negotiation, not by how hard it is to fix. Each item names what a buyer reads into it and the shape of closing it.

Your code never leaves

Zero dependencies, zero network calls. The engine runs on your machine, under your control, before you've decided to trust anyone with the repository.

What it examines

The four questions a buyer will ask about your code. Plus the one about your habits.

Is the AI real?

Your prose is scanned for AI claims, then your code is checked for the model integration that would make them true — including whether it's reachable from your product's entry points. If your marketing has drifted ahead of your code, you want to be the one who finds out.

Who holds the system?

Truck factor and single-owned critical files, computed from your commit history. Buyers price this at a 0.5–1.5x multiple reduction — and it's the finding with the longest lead time to fix.

What's in the license graph?

Copyleft obligations that attach to proprietary code. Counsel resolving this on your schedule is an engineering decision; counsel resolving it in a data room is a deal issue.

What leaked into the repo?

Credentials committed to your repository, redacted in the report. Rotate them now and you can show a buyer the rotation dates — that's a different conversation from deleting the line.

Plus code-quality signals: whether tests and CI exist, and which files are simultaneously high-churn and single-owned — the maintenance cliff a buyer's engineering lead will find on day one.

How it runs

Screen, close, re-screen.

The readiness screen

Run the engine against your codebase — on your hardware if you prefer. You get the punch-list, ordered by what it costs you, with the evidence a buyer would see.

Close what matters

Work the list on your own timeline. Most items are engineering tickets when they're not deal terms — knowledge transfer, credential rotation, a licence decision, copy that matches the code.

Re-screen before the process

Run it again and get a delta: what changed, what's gone, what's new. Walk into diligence knowing what their screen will say.

What this is not: Assay Ready doesn't price your business, and closing every item guarantees no valuation outcome. It doesn't replace counsel on licence or disclosure questions. It reports what a static pass over your repository can establish — and says plainly what it did not assess.

The findings are coming either way. Choose when.

Tell us where you are in the process. If a readiness screen isn't the right instrument yet, we'll say so.

Book a readiness screen