Launching today
Clippy
Parallel AI contract analysis. Compare the findings.
1 follower
Parallel AI contract analysis. Compare the findings.
1 follower
Compare AI analyses of a contract side by side. Find clauses and questions worth discussing with a qualified reviewer. MIT code. AI analysis, not legal advice.
Free
Launch Team
Render Sandboxes: a computer for every agentIsolated compute, persistent state and a real filesystem
Promoted
1/4 · A confident answer isn't always a careful reading, and that matters when the document is a contract.
Clippy lets you upload a contract, choose AI models and compare their analyses side by side. I built it to make disagreements, omissions and useful questions easier to spot before you talk to a qualified reviewer. Seeing where models read the same clause differently can show you which questions to raise first. It is a way to prepare for the conversation, not a replacement for it.
It isn't legal advice, and several models agreeing doesn't make their answer accurate. The value is arriving with a better brief: the clauses you're worried about, the assumptions to check and the questions worth spending human time on.
This is Day 5 of 11 of the Product Hunt Marathon: 11 free, open-source launches, one a day, 10-20 November.
The code is MIT-licensed. Before uploading anything confidential, consider which providers will receive it and what you owe the other parties to the contract.
What would make this comparison genuinely useful in your review process, rather than just another long AI answer?
2/4 · Inputs, privacy and review costs
Upload a contract, choose models and compare their analyses side by side. Supported inputs include PDF, DOCX, TXT and Markdown. I want the comparison to make useful questions easier to find.
This is not legal advice, and a model panel is not a substitute for a qualified lawyer. Before uploading sensitive documents, review the selected providers' data handling. OpenRouter inference costs depend on the models and usage.
Use Clippy: https://clippy.legal
Source and issues: https://github.com/paulfxyz/clippy
Campaign: https://ph.paulfleury.com/
3/4 · Why the human handoff matters
The spirit I want to carry forward is the right to inspect and question systems that affect us. Here that means making the clause and competing interpretations visible, not replacing professional legal judgment with a confidence score.
Today's launch is therefore about the handoff. Can a small tool help you arrive at an expert conversation better prepared, rather than pretending to eliminate that conversation?
A CHO-led team could work with legal and operations to turn repeated contract-preparation tasks into one maintained workflow. The goal is not to replace legal judgment, or simply replace a SaaS bill with unlimited model calls.
It is to improve the handoff: ordinary code for predictable preparation, model analysis where it adds value, and a qualified reviewer for the decision. Reducing redundant full-panel runs is an improvement direction, not a claim that Clippy already optimises them. Evaluate the quality and full cost of a completed review, including the human work.
CHO (Chief Hacking Officer) is a deliberately playful title for a serious role: I'm open to a CHO brief of up to 30h/week with a great company or startup, a worthwhile product, trust and a modest budget. Job offers: paulfleury.com/contact.
4/4 · Help improve the comparison
I would value a non-sensitive example where an important clause is missed, an explanation is misleading or the side-by-side reading is difficult to follow. Those give us concrete acceptance criteria for another iteration.
The code is MIT-licensed. If you extend it with AI, constrain the work to a particular comparison problem, decide the budget first and test against the original document. Review the code and data handling before contributing a PR. Commercial reuse is allowed under the licence with its notices retained.
I plan to weigh public feedback by usefulness, risk and cost, rather than treating every request as another feature to bolt on.
Web Summit Lisbon is over, and I'm grateful to the people I met there. I live in Lisbon and am happy to meet makers or backers around the city: X (@paulfxyz) or paulfleury.com/contact.
Tomorrow is Day 6 of 11: Mercury.
A few other things, briefly, none part of the marathon. Openline (openline.com) is my main venture and startup: it aims to reshape the eSIM market, with a deep pipeline of ideas ready to execute, and I'm raising, so VCs, angels and investors are welcome to DM me. Iris (iris.online) and Digital Spot (digital.spot) are spin-offs of tools built for Openline, and I'm looking for investors and early customers for both.
Enki (enki.ngo) is rather a thought experiment, a non-profit, local-first AI public good too big for me alone; I'm looking for philanthropists, donors and helpers. Cup of News (cupof.news) is a Sunday-afternoon hobby that is soon turning into an app, and I'm looking for contributors and users.
Reach me on X (@paulfxyz), LinkedIn or paulfleury.com/contact.