I'm building an AI contract tool for founders. Tear it apart.
I've been commenting on this forum for a bit and wanted to finally share what I'm building and get honest feedback.
Quick background. I ran a product brand for almost two years. Built it from scratch with no legal background and no budget for a lawyer. Over 18 months I wrote 80+ contracts across factories, engineers, scientists, vendors, agencies, and contractors.
I once hired an influencer for 20 reels across different outfits and locations. They delivered 2 outfits and 2 locations. Technically fulfilled the contract because I never defined what "different" meant. That was on me.
That experience plus 79 others taught me that most founders don't lose because they had no contract. They lose because the contract was vague, missing key clauses, or written for someone else's situation.
So I built Claustar.
It's an AI-guided contract creation platform that asks about your situation before writing a single clause. Attorney-reviewed templates under the hood. Jurisdiction-aware. Plain-English output. Built to stay clearly on the document preparation side of the legal line, not to give legal advice.
Launching next week.
Before I do, I genuinely want the unfiltered reaction from this community.
Does the idea resonate, or does it sound like something you'd never actually use? What would make you trust it with a real agreement? And is there anything about how it's positioned that feels off?
Not looking for encouragement. Looking for the honest reaction you'd have if a friend sent you this.
Hardest part here is not generating contract text, it's earning trust when the edge cases show up. For founders, the product that feels useful is usually issue-spotting, plain-English explanations, and a clear handoff to a real lawyer, not a magic "safe contract" button. Would pressure test where you draw the line on liability, jurisdiction, and bad inputs before adding more AI polish.
The influencer contract story is painfully relatable. When you've been through enough of those "technically fulfilled" moments, you start to realize that the expensive part of contracts isn't the lawyer — it's what you didn't think to include.
Honest feedback on positioning: the "AI-guided" angle is strong, but I'd lean even harder into the specific pain you described — the vague contract that technically gets fulfilled but screws you. That story is way more compelling than "AI contract tool" because every founder has their own version of it. When we were building our SaaS, we had a similar realization with our own product messaging — the specific failure scenario always outperformed the generic feature description.
One thing I'd want to know as a potential user: how does Claustar handle the "I don't know what I don't know" problem? The influencer contract story is a perfect example — you didn't know to define "different" until it was too late. Does the guided flow surface those kinds of blind spots proactively? If it does, that's the real value prop, not just "write a contract faster" but "catch the clause you didn't know you needed." That's the kind of thing that would make me trust it over a template I downloaded from the internet.
Really appreciate you coming back with more. The "catches the things you'd miss" framing is sharper than what I had. You're right that the influencer story isn't about needing a lawyer. It's about not knowing the question to ask. That's actually the whole intake flow. Claustar asks the questions you wouldn't think to ask before it writes anything.
On the "I don't know what I don't know" problem directly. Yes this is baked in. The guided questions surface standard clauses for your specific agreement type that most people skip. You don't need to know what a kill fee is for it to show up when you're hiring a contractor. It surfaces it and explains why it matters.
On attorney-reviewed framing. Taking that note seriously. The intent was credibility not replacement. Will be tightening that language before launch.
On segment. Starting with product founders and freelancers. People who are signing agreements with contractors, vendors, and agencies regularly. Not employment lawyers. Not enterprises. The retail brand experience I had is the exact ICP.
that's the right call. 'you're always clicking send' is a sentence that will close a lot of deals once founders internalize it
The influencer contract story really hits home. I'm building an AI tool in a completely different space (ad creative generation) and one of our earliest lessons was the same — ambiguity kills you. We had a freelance designer who delivered "social media assets" that were just cropped versions of one image. Technically met the brief, but useless in practice.
What I think is really smart about your positioning is staying on the "document preparation" side vs. legal advice. That's the exact line that makes AI tools trustworthy in regulated-adjacent spaces. One thing I'd push on: how are you thinking about the moment a founder finishes generating a contract? That post-generation step (review checklist, flagged ambiguities, "here's what a lawyer would double-check") could be a massive trust builder and upsell opportunity. Would love to see how launch week goes — rooting for you.
Strong problem selection — the influencer example makes the risk concrete fast.
If you want more conversion from skeptical founders, I’d test this structure on the landing page:
That combo usually increases trust because it shows process, limits, and evidence in one scan.
If useful, I put together a quick 5-point homepage teardown rubric you can run in 10 minutes: https://roastmysite.io/?src=external_manual_nonhn_ih_ai_contracttool_tearapart_20260328_c1_usd_presell_hv
The three things you laid out are going directly into the pre-launch checklist. Outcome framing, credibility stack, risk control. The "when to hire a lawyer instead" decision box is something I hadn't thought about explicitly but it's the right move. Shows we're honest about the product's limits which builds more trust than pretending it covers everything.
trust is the real moat here and I think the comments are right. same thing happened with AI tools in my PM work — teams adopted summarization and sprint planning tools fast once they saw legal own the contracts rather than AI generate them. the workflow that worked: AI drafts, human reviews, human clicks submit. nobody wants a vague contract any more than a PM wants an AI autonomously closing tickets. what does the approval UX look like in your tool?
The approval UX is exactly what you described. AI drafts, you review every clause in plain English, you make the final call before anything gets sent. The platform never decides for you. That distinction is also the legal line we're careful to stay on. You're always the one clicking send.
The influencer story is a perfect encapsulation of the problem. Vague contracts fail at the edges, and founders only discover the edges when things go wrong.
The approach of asking context before writing is the right call. Most legal tools start from templates and make you adapt to them. Flipping that means the output actually reflects the situation instead of just looking official.
One thing worth pushing on: what happens when the founder does not know what they do not know? The influencer example is obvious in hindsight. But a lot of contract failures come from missing a clause the founder never thought to ask about. Does Claustar surface those gaps proactively?
Yes Claustar surfaces those gaps proactively. That's the whole point of the intake flow. You don't have to know what a revision limit clause is for it to come up when you're hiring a designer. The flow asks about your situation and surfaces the clauses that apply. The influencer example is actually built into the deliverable definition question now because of exactly that experience.
The positioning makes sense, but I think trust is the whole game here. For most founders the question will not be "can AI draft this?" but "where could this fail in a way that hurts me later?" I would lean hard into showing boundaries: what the product is good for, what situations should trigger human review, and maybe a clause-by-clause explanation of why something was included. That makes it feel less like a black box and more like a guided drafting tool. If you nail that, the value proposition gets much easier to believe.
This is a really strong problem to tackle — especially the “contracts fail because they’re vague” part.
One thing that stood out to me: trust will probably be your biggest challenge, not the product itself. When it comes to contracts, people are extremely risk-averse.
Maybe worth thinking about:
The positioning feels clear, but I’d lean even more into “reducing risk” rather than just “making it easier”.
Curious — are you planning to let users tweak clauses manually after generation or keep it more guided?
Yes to manual editing. You can tweak any clause after generation. The guided flow is the starting point not the final word. You own the document. On showing under the hood. Working on exactly this before launch. Clause-level explanations for why something was included are already in the product. Making that more visible on the landing page based on this thread.
Honestly, dealing with clients myself, I would not use it to create my whole contract but i would definitely use it to refine my contract, clear my loopholes and give me insight on potential clauses that i may be missing. If your AI tool has this time of feature instead of primarily just drafting its own contracts then I personally think it is powerful and yes, I would use it!
That's actually a really useful data point. The refine and loophole check use case is something several people here have mentioned. Worth understanding whether that's a separate workflow or the same product framed differently. Would you be open to a quick conversation about how you currently use contracts and what that refinement process looks like for you?
No problem. I would love to chat about it, just send me a quick message on my Linkedin. I am not eligible for links yet but just search Boipelo Nsimbini and you will find me.
Hi
I provide eligible AI businesses access to GCP/AWS credits for the next 24 months to help reduce infrastructure costs.
If this is relevant, happy to share more details.
https://www.linkedin.com/posts/sai-rithvik-2176302b1_eligible-ai-companies-can-access-up-to200k-activity-7442865181254209536-EiDB
Interesting concept! As a solo founder myself, contract management is one of those things I keep putting off. How do you handle data privacy for sensitive legal documents? I'm building an E2E encrypted secrets manager and security is always top of mind.
Good question and one I expected here. Documents are encrypted at rest and in transit. We don't sell or share your data. Nothing you upload is used to train models. Given the nature of what people are putting in these documents, security wasn't optional. Happy to go deeper on the architecture if useful.
Hi
I provide eligible AI businesses access to GCP/AWS credits for the next 24 months to help reduce infrastructure costs.
If this is relevant, happy to share more details.
https://www.linkedin.com/posts/sai-rithvik-2176302b1_eligible-ai-companies-can-access-up-to200k-activity-7442865181254209536-EiDB
This solves a real pain — I’ve messed up contracts myself just because things weren’t clearly defined.
If you can make users feel safe (clear limits + review step), this could be something founders actually trust and use.
Just checked it out. Clean design. My one piece of feedback would be to show a real example on the landing page. Like a before (messy contract) and after (clean AI generated one). People need to see the magic before they sign up. Otherwise looks solid, keep shipping!
Before/after is going on the landing page. Multiple people in this thread said the same thing and they're all right. You shouldn't have to sign up to see what the product actually does to a contract. Working on getting a real example up before launch.
The influencer story is painfully relatable — those "technically fulfilled" loopholes are where most founder contracts fall apart. The idea definitely resonates. I've been on the other side of this building an AI tool for ad creatives, and the trust question you're asking is the right one to obsess over.
For what it's worth, the thing that moved the needle for us on trust wasn't more features or disclaimers — it was showing the work. When users could see WHY the AI made a specific choice (not just the output), they trusted it enough to actually use it. For contracts, I imagine that's even more critical. If the tool says "you should include a non-compete clause," founders need to understand the reasoning, not just accept it.
One honest reaction: the "attorney-reviewed templates" positioning is smart but might also be a double-edged sword. It sets an expectation of legal reliability that could scare some people away if they're worried about edge cases. Have you thought about framing it more around "this catches the things you'd miss" rather than "this replaces your lawyer"? The pain point in your story wasn't that you needed a lawyer — it was that you didn't know what questions to ask. That framing feels stronger to me.
This is exactly the kind of build I needed
to see. Just started my own daily build challenge
— shipping 1 AI tool per day . Day 1 was rough
but learned a lot about rules the hard way!
Hi
I provide eligible AI businesses access to GCP/AWS credits for the next 24 months to help reduce infrastructure costs.
If this is relevant, happy to share more details.
https://www.linkedin.com/posts/sai-rithvik-2176302b1_eligible-ai-companies-can-access-up-to200k-activity-7442865181254209536-EiDB
— the influencer contract example is the kind of specific, painful scenario that makes people immediately get why this needs to exist. As a fellow AI tool founder, one thing I'd push on: how are you thinking about the trust gap? Founder— maybe show exactly which clauses came from templates vs. which were AI-generated. Transparency about what the AI is and isn't doing could be a real differentiator against generic "AI writes your contract" tools. Also curious about your go-to-market. Are you targeting a specific founder segment first (e-commerce, agencies, freelancers) or going broad? In my experience building AI tools for small businesses, the narrower you go early on, the faster you find people who will pay on day one. Either way, the problem is real. Bookmarking this to check back after launch.
Blunt take: idea is solid, but trust conversion is where this wins or dies.
If I were in your shoes, I’d tighten 3 things before launch:
If useful, I can do a fast conversion teardown of your homepage + signup flow and point out the top leaks before you push launch traffic:
https://roastmysite.io/go.php?src=ih_claustar_tearapart_20260327_1325
Hi
I provide eligible AI businesses access to GCP/AWS credits for the next 24 months to help reduce infrastructure costs.
If this is relevant, happy to share more details.
https://www.linkedin.com/posts/sai-rithvik-2176302b1_eligible-ai-companies-can-access-up-to200k-activity-7442865181254209536-EiDB
Love this ‘tear it apart’ is the right mindset 👏
I help founders test their apps and provide structured feedback on bugs, UX friction, and improvement opportunities.
I can run a proper test and give you a clear, actionable report if you want happy to help!
Really interesting! Claustar reminds me a bit of EqualDocs - both help founders avoid contract mistakes with AI guidance. Claustar focuses on drafting clear contracts for one party, while EqualDocs also adds multi-party negotiation and risk flags. Love seeing different approaches to the same problem!
Aware of EqualDocs. The core difference is where the product starts. Claustar starts before the document exists. Situation-first intake, then generation. The negotiation layer is coming but the creation problem is what we're solving first because that's where most founders get hurt before anything ever reaches a counterparty.
Got it thanks
Tools like this rely a lot on clarity and trust, especially since users are dealing with contracts.
Even small interaction details (like how steps flow or how actions confirm) can make a big difference in how reliable the product feels.
Interested to see how your current flow works.
Agree on interaction details. The flow is guided step by step so there's no moment where you're staring at a blank document wondering what to do next. Would genuinely value your eye on it before launch if you're interested.