Founders Simon Horsman and Daniel Wood are putting Quilty's screenplay analysis inside two chat assistants, with tools to compare a rewrite against its prior draft.
By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)
· Published
Primary source: [PR Newswire](https://www.prnewswire.com/news-releases/quilty-brings-screenplay-coverage-into-claude-and-chatgpt--and-scores-the-rewrite-against-the-last-draft-302901887.html)
Why it matters #
Quilty is betting that screenplay analysis becomes more useful when writers can consult it during revisions. Its new integrations make the feedback easier to reach; they do not establish that a higher score means a better script.
Simon Horsman and Daniel Wood are bringing Quilty's screenplay coverage into Claude and ChatGPT, giving writers a way to query existing reports and compare a rewrite's score with the previous draft. The integrations, announced on October 8th, put Quilty's analysis inside chat tools writers already use, rather than leaving its reports as PDFs to consult separately. Quilty's announcement says writers can connect an account and ask about story notes, character work, market analysis and talent recommendations.
The move fits the founders' unusual pairing. Horsman is an entertainment producer and lawyer who worked as general counsel at PriceGrabber.com before moving into film and television financing. Wood studied Cinema-Television at USC, co-founded PriceGrabber and later returned to independent film production. Their shared history gives Quilty a commercial-technology lineage and a production-world pitch: apply software and data analysis to decisions filmmakers often make with limited staff and money. Quilty's March launch described it as a broader entertainment-intelligence platform; this integration brings its screenplay analysis into the tools where writers can act on that feedback. Quilty's launch announcement said the pair had worked together at PriceGrabber, which Experian acquired in 2005.
Wood has framed the product around giving independent filmmakers information they typically cannot afford to assemble themselves. In a September interview with Film Threat, he described Quilty as a way to use AI to help indie filmmakers assess scripts and projects, rather than hand creative control to a model. Now the wager is that writers will return to analysis during revisions if they can reach it through a conversational assistant.
A report becomes part of the rewrite loop
The central feature is draft-over-draft comparison. After a new version has been analyzed, a writer can ask Quilty to compare it with the earlier analysis and show changes in the overall Quilty Score and its component pillars. That turns the score from a one-time readout into feedback a writer can consult while revising. Whether writers find a score useful as a revision target is a separate question from whether it is convenient to retrieve.
Quilty says its score combines four weighted categories: Story & Craft, Commercial Viability, Cultural Resonance and Production Reality. Quilty describes the current version as a 0-to-100 measure divided into 15 sub-dimensions, with a confidence interval and a model-agreement figure. Quilty says an analysis runs through 33 stages and can produce more than 70 types of report sections. Those are Quilty's descriptions of its system; the announcement does not offer independent validation that higher scores reliably correspond to stronger scripts or better commercial outcomes.
The connection is designed to reveal information gradually. Quilty says 13 read-only tools let an assistant start with the score and a short brief, then retrieve specific sections, scenes, talent recommendations or comparable-film research. Its Claude connector documentation describes access to completed analyses, scene search, score and pillar breakdowns, comparisons between analyses, and a temporary link to the full PDF. The tools are read-only: the documentation says Claude cannot upload scripts, initiate a new analysis, change account settings or spend credits. The October announcement also names ChatGPT; Quilty describes the same general workflow there.
Quilty says identifying information is removed before scripts reach its scoring models, and describes the score as a diagnostic assessment rather than a box-office forecast. It also says it routes different tasks among models from Anthropic, OpenAI, xAI, Google, Perplexity and DeepSeek. That is Quilty's account of its process, not evidence that a collection of models removes the uncertainty in assessing creative work.
The score still has to earn trust
Quilty faces a practical test beyond making reports easier to reach: writers must believe the feedback helps them make better decisions. In April, TheWrap ran Quilty on four known screenplays: Christy, Sinners, Barbie and Die Hard. Its test gave Christy the highest overall score and found parts of the creative analysis off base, while noting that the tool was more useful for budget breakdowns and production issues. Quilty's blind-scoring design was intended to prevent a film's reputation from influencing its assessment, but the results showed the limits of treating a composite score as a proxy for creative judgment.
That caveat puts more weight on the rewrite comparison. A score change can tell a writer how Quilty evaluated a new draft under its framework. It cannot, by itself, establish that the rewrite is more moving, more original or more likely to find an audience. Quilty explicitly says its score is not a box-office prediction, and the test by TheWrap offers a concrete reason to keep that distinction in view.
Horsman and Wood are selling access to structured development feedback, not a machine verdict on whether a screenplay deserves to be made. Putting that feedback inside Claude and ChatGPT may make it easier to use between drafts. The harder product challenge is persuading writers that the number is a useful signal, and that its specific notes deserve attention when they conflict with a writer's own judgment.