Defending · Accused infringer
Infringement is conjunctive — the plaintiff must show every limitation present. Non-infringement is disjunctive. Contist works an asserted chart element by element, surfaces the strongest column instead of averaging the weak ones, and grounds the prosecution-history arguments the plaintiff never raises against themselves.
Why a defense chart is not a plaintiff chart
A plaintiff chart wants its weakest cell shored up, because under the all-elements rule a single missing limitation defeats the claim. A defense chart wants that cell found and led with. Tools built for the asserting side aggregate toward a verdict on the claim; what you need is the one element where the record is strongest, named and separated from the noise.
The arguments differ too, and they do not share a scale. “The product lacks this element” is answered with summary judgment. “Your contention is conclusory” is answered with interrogatories. “The claim is invalid” is a different case entirely. Contist assesses each of those axes from only the findings that speak to it, so a contention defect can never inflate a technical call.
The argument the plaintiff will not make
Every narrowing amendment made for a reason related to patentability raises a presumption that the territory between the original and amended claim was surrendered. That history is sitting in the file wrapper, and it is the patentee's own words.
Read: Prosecution History as a First-Class AssetHow it works
Contist takes the contention you were served, rebuilds it as a structured grid, and runs each limitation against the accused-product record and the patent's own prosecution file.
The served contention is parsed into claims and limitations and checked structurally — split elements, broken labels, missing limitations, duplicates. You correct the parse before anything is analysed against it.
Your own technical documentation, source, and specifications. The plaintiff's cited exhibits are surfaced read-only and never indexed beside it — they were curated to show infringement, and they would skew every technical angle toward finding the element present.
The patent's full prosecution history from USPTO ODP, resolved claim by claim: what each amendment actually changed, what the applicant argued to get it allowed, and which claims were narrowed, when, and why.
Eleven angles across non-infringement, contention sufficiency, and invalidity — technical presence, mechanism, claim construction, prosecution estoppel, priority, family, prior art. Each one reports per limitation, or reports that it could not run.
Attorney positions are written inline, per limitation, and travel as a first-class column into the exported chart. The internal copy carries the calls and the analysis basis; the served copy drops all of it.
Capabilities
The file wrapper is resolved into claim state — what each paper restated, what it carried forward, what status the examiner assigned — so comparisons are between claims, not between documents. Where a claim narrowed, the amendment is read element by element against the rejection it answered and the remarks filed with it, and the direction is derived from what changed rather than guessed at holistically.
Under Festo, a narrowing amendment made for a reason related to patentability raises the presumption of surrender. Under Warner-Jenkinson, an unexplainednarrowing raises it too — silence favours estoppel, not the patentee. Every finding carries the added limitation and the applicant's own argument, quoted.
The call on each limitation is the non-infringement axis alone. Contention sufficiency and invalidity are assessed and reported separately, because they are different arguments with different remedies. Letting them share one scale is how every element in a chart ends up flagged “priority” with not one of them led by a technical finding.
“No attack found”, “the evidence is adverse” and “we do not know” are different answers and stay distinguishable from the analysis through to the export. An element the record contradicts must not read as silence, and an angle that never ran must not read as a clean result.
Which angles ran, which did not, and why — stamped on the run and printed into the exported chart. Naming only the gaps lets a reader assume the rest was a full analysis.
Where the served chart asserts an element without mapping it to anything, that is a discovery problem, not a technical one — tracked on its own axis with its own remedy.
The export is work product unless you say otherwise. The internal copy is labelled in the filename; the served variant drops every call, strength and basis. The failure worth guarding against is sending the wrong file.
Limits, stated up front
Every call is a proposal. Narrative and export only run on what counsel has rated, and editing a limitation's text marks the analysis stale rather than leaving a verdict standing against words that no longer exist.
A failed comparison, a claim missing from an OCR'd listing, or an angle whose input does not exist is reported as unassessed — never as a clean finding. A confident wrong answer is the most expensive output this product could produce.
Prosecution history is the richest signal available before claim construction, expert discovery, and newly found art resolve. It is not an oracle, and the chart says where its reasoning stops.
We work directly with a small number of litigation teams during the design partner program. Send us the patent and the contention — we will come back with the prosecution history and the angles it opens.