Tell us the situation. Get a playbook in 90 seconds.

The starting line is a single sentence in your own words. A $6,300 ER bill where the doctor was out-of-network. A salary offer that's 12% below market. A lease renewal that jumped $300. A debt collector calling about a credit card you already paid. Whatever it is, type it the way you'd describe it to a friend.

You can also paste the bill or contract text directly. The system reads what's there and pulls out the fields that matter — the dollar amount, the counterparty, the dates, the line items. You don't fill out a form. You don't pick a category from a long dropdown.

From here, you have 14 categories the system knows deeply — medical bills, salary & offers, telecom, rent & housing, debt & collections, vehicles & lease, insurance, freelance rates, used goods, jewelry, travel, retail, services, and collectibles. It routes your situation to the right playbook automatically. If your situation doesn't fit cleanly into one of those, it picks the closest and tells you so.

This is the step most generic AI tools skip — and it's the step that makes the rest of the brief actually useful. Without it, an AI gives you a generic checklist that could fit anyone. With it, the brief is tuned to your numbers, your state, your leverage points.

The questions are short and specific to the category. For a medical bill, we'll ask whether you have insurance, what state the procedure happened in, whether it was emergency or scheduled, whether you've already received an Explanation of Benefits. Each question takes seconds, and the question after it is informed by your prior answer — so you never see an irrelevant question.

The reason these questions matter: they determine which specific federal protection applies (the No Surprises Act covers emergency out-of-network billing, but not scheduled), which letter template fits, what tactics the hospital is likely to push back with, and how strong your walk-away position is. The brief that comes out the other end is built on these answers.

If you don't know the answer to a question, you can skip it. The brief will say what changes if it knew.

About 90 seconds after you finish the clarifying questions, a brief appears. It's a real document — sections, headers, citations — not a chat thread you have to scroll through. You can read it in five minutes. You can print it. You can take it into the conversation.

Here's what's actually in it:

The brief is generated specifically for your situation using Claude Sonnet, with cached prompts tuned per category so the outputs stay consistent run-to-run. Every cited law links to the original source — the eCFR for federal regulations, state legislature sites for state statutes, government agency sites for guidance. If we cite a price, we link the comp.

When the brief can't find a citation for a claim, it marks the claim as “uncited insight” instead of fabricating one. Hallucinated citations are the most common way AI legal tools go wrong, and the most common cause of FTC action against them. We'd rather a brief say “we don't have a citation for this” than make one up.

The brief is the artifact. What you do with it is up to you. Most people read it once, skim the script, and pick up the phone. Some people use practice mode first.

Practice mode (optional)

Practice mode is a roleplay where the AI plays the counterparty — the billing rep, the hiring manager, the landlord — and you run through the conversation. You can try the opening three different ways, see how each one responds, and pick the version that fits your voice. Most negotiations get won or lost in the first 30 seconds. Practice mode is where you make sure your first 30 seconds is strong.

We don't record audio of any real call with a real counterparty. Practice mode is between you and the AI only.

Run the negotiation

This part is on you. We don't call the company for you, don't email on your behalf, don't represent you. You make the call, send the email, walk into the room. That's a deliberate choice — having you in the room means you control the conversation, you read the other person, you decide when to walk away. The brief is your prep, not your proxy.

If you want a quick read on something mid-conversation, you can open the brief on your phone and there's a live coach chat that can answer follow-up questions. “They just said X — what should I say back?”

Log the outcome

When you're done, come back and log what happened. Won, partial, lost, no action, unclear. If you saved money, we charge you $14.99. If you didn't, we don't. You can come back at any time — there's no deadline.

Outcomes are anonymized and rolled up into the public Track Record. Your specific situation never appears; only the category, the outcome, and the savings amount, and only after at least 50 outcomes have been logged in that category. Your numbers are how we earn future skeptics' trust.