Agents that run diligence with you
Not a chatbot. Each agent is a workspace in your deal room — a dashboard, a chat, and the tools to review and correct what the AI extracted.
Meet the agents
Each agent is a workspace for a distinct part of diligence — pick one to see how it works in the room.
Due Diligence Agent
The Due Diligence Agent turns a pile of uploads into a reviewed data room. It auto-sorts documents into folders, maps them to a diligence checklist across corporate, financial, legal, tax and IP domains, and shows exactly what’s covered and what’s still missing. Ask it anything and it answers grounded in your files — every citation opens the source page. Map or unmap a document by hand and coverage recomputes instantly.
- Checklist coverage Maps documents to a mandatory / recommended requirements checklist by domain, and flags what’s still missing.
- Auto-categorised documents Sorts uploads into folders and tags each file’s type, so the room is organised from day one.
- Grounded Q&A with citations Answers diligence questions from your files — every citation opens the exact source page in a drawer.
- You stay in control Map or unmap a document by hand and coverage recomputes — human-in-the-loop, not a black box.
Yes — a signed Founders’ Agreement with 4-year vesting and a 1-year cliff, equity split 55 / 30 / 15.
Financial Analyst Agent
The Financial Analyst Agent reads your statements and financial model and reconciles them into one normalised set of figures — revenue, EBITDA, margins, ARR, burn, runway and the ratios reviewers check first. It shows the source and a data-quality flag for every number, flags conflicts across documents, and lets a human override any figure or confirm a line-item mapping. Every change recomputes the derived metrics and lands in an audit trail.
- Reconciled statutory figures Extracts and normalises P&L and balance-sheet figures into one dashboard of KPIs, trends and ratio health.
- Conflict detection Flags where the same metric disagrees across documents — e.g. audited revenue vs. the model’s run-rate.
- Human-verified, with an audit trail Override any figure or confirm a mapping; derived metrics recompute and every edit is logged.
- Traceable to the source line Every figure links back to the exact statement line or page it was read from.
ARR $18.2M, ~16.5 months runway — but audited revenue ($49.8M) conflicts with the model ($52.9M).
Answers you can defend
Diligence runs on trust. Every agent is built so a reviewer can check the source behind any answer.
Grounded in your room
Every answer is drawn only from the documents inside the deal room — never the open web.
Always cited
Each claim links back to the exact source document, page, or slide so reviewers can verify.
Human-in-the-loop
Correct any figure or document mapping by hand — coverage and metrics recompute, and every change lands in an audit trail.
Put the agents to work on your next deal.
Spin up a secure room and let AI agents field the diligence questions — grounded in your documents, with citations.