412 FILES INDEXED · 9 GAPS OPEN
016 agents
Data and completeness
A databook that ties to the trial balance, and a written list of what is still missing.
data room indexer · file classifier · trial balance tie-out and 3 more
The agent layer for
transaction services.
Agents that read the data room, build the analysis and draft the report. Your team reviews and signs.
Who it is for
A new way to run a diligence.
Agents change how the work gets done, not who does it. The same way of working holds whatever the size of the team.
BIG 4 DEAL TEAMS
Inside a large deal team
Agents carry the tie-outs, the chases and the exhibits, so every workstream arrives at review already evidenced.
INDEPENDENT ADVISORY
Inside a small advisory firm
The same scope, the same audit trail, the same standard of report, run end to end by the people who signed it.
One engagement
Start to finish, the way your team already runs it.
Project Atlas, a fictional buy-side engagement. Five stops. At each one the agents bring the work, and someone on your team makes the call.
- 1
Data room
412filesIndexed and classified. Fourteen gaps drafted for the request list.
Yousend the request list
- 2
Accounts tied
411of 412 linesTrial balance tied to the databook. One exception, explained.
Youaccept the exception
- 3
Earnings quality
Twenty-three adjustments. Seventeen accepted against your policy, six held for you.
Youdecide the six held items
- 4
Draft report
Sections in your template, findings ranked by price impact, every figure linked.
Youedit and approve each page
- 5
Sign-off
PartnerNothing reaches the client before a named reviewer signs.
Yousign
Watch it run
One engagement in under a minute, screen by screen, from the overview to the answer the client reads.
The app
Checked before it reaches you.
Each adjustment arrives with the checks a senior reviewer would run. You see what passed, and the one point that needs your judgement.
Each proposal comes with its source and its checks
Get the desktop app for macOS and Windows. Already working in Claude, Codex or Copilot? Connect Jony to it.
Example from the app: an add-back for an owner bonus above market rate, plus 0.42 million euros to EBITDA. Four checks passed: tied to the trial balance at cell C214, compared with market pay, the same pattern from FY22 to FY24, and not counted again in net debt. It sits above the 0.25 million euro threshold set in your policy, so a person decides.
The layers
What sits under a report.
Models at the bottom, your signature at the top. Each layer does one thing, and each one can be checked.
- 01
Models
Frontier models from Claude, OpenAI or Grok, on your own subscription. Swappable, never a lock-in.
- 02
Harness
Planning, tools and tests around each model call. One agent, one job, one output, checked before it lands.
- 03
Your data
The data room, the trial balance, the databook. It never reaches Jony and never trains anything.
- 04
Your method
Your report template, your databook conventions, a few past reports. Encoded once, followed on every deal.
- 05
The agents
Families of agents: data room, earnings, balance sheet, report. Each one writes to your method.
- 06
Your review
Scope, open items, every number traced to its source cell. Judgement and sign-off stay with your team.
The agents
Dozens of agents, one job each.
A diligence report is a few hundred small jobs: a tie-out, a peg, an exhibit, a chase email. The agents come in four families, one for each part of the work.
REPORTED 16.2 → ADJUSTED 16.6
027 agents
Earnings quality
The bridge from reported to adjusted EBITDA, with every line tied back to a source cell.
EBITDA bridge builder · add-back adjudicator · run-rate and annualisation and 4 more
peg
12-MONTH AVERAGE · PEG 4.1M
036 agents
Balance sheet and cash
A peg with twelve months of seasonality behind it, and net debt with each debt-like item named.
NWC normalisation · seasonality and peg · debt-like classifier and 3 more
6 SECTIONS · 41 EXHIBITS
046 agents
Report and delivery
Your report, in your template, in your order of priority. Every number stays traceable to its source.
findings ranker · section drafter · exhibit builder and 3 more
databook buildertrial balance tie-outadd-back adjudicatorEBITDA bridgecohort analysismargin walkrun-rate & annualisationrelated-party screenNWC normalisationseasonality & peg
debt-like classifierrestricted cash screencapex & lease reviewEBITDA to cashsection drafterfindings rankerexhibit buildercross-reference checkerIRL managermanagement call prep
Data handling
Client data never reaches Jony.
Diligence material is the most confidential data a firm handles. The product is built around that.
Beta firms get the data-handling note, the model setup and a walkthrough of what is logged, before any deal data is loaded.
COMMITMENTS06
- Figures computed on your computernothing is uploaded to Jony
- Bring your own model subscriptionno tokens resold
- No client data used for trainingJony never receives it
- Every figure traces to a source celldatabook reference on every line
- No output leaves without a reviewera person signs, not the model
- Built by people who ran the worktransaction services, M&A, investment
Formal certification is on the roadmap, not claimed today.
Why we built it
We did this job for years. Building the databook took days. Reading it took a morning.
So we built agents to do the building. They follow your method, every figure they write points to its source, and the person who signs still decides.
Questions
What partners ask first.
01Does this replace the analyst?
No. It takes the reformatting, the tie-outs, the file chasing and the rebuilt schedules. Judgement, client calls and sign-off stay with the team.
02Where does our client data go?
It never reaches Jony. In the app and the plugin, the agents run on your computer with your own model subscription; with the connector, they run in your own assistant. Either way, what the assistant reads goes to its provider, under your firm's own agreement. We do not hold deal data and we do not train on it.
03How do we know a number is right?
You can open it. Every figure carries its source cell and every adjustment carries its document. Anything that rests on a judgement call is flagged.
04How granular does it get?
Down to the artefact. There is no single diligence agent. There is a tie-out agent, a peg agent, an exhibit builder, a chase drafter. Each has one job, one output and its own tests.
05Will it work with our report template?
Yes. We take your report structure, your databook conventions and a few past reports, and the drafting agents write to them.
06What does the private beta involve?
One live engagement, run in parallel with your normal process, so you can compare the output. We take a few firms at a time and work with the team on the first deal.
07How can we work with Jony?
In three ways. The Jony app runs on a Mac (Apple silicon, macOS 14 or later) or a Windows PC (Windows 10 or 11, Intel or AMD). It computes the figures on your computer with your firm's Claude plan (a ChatGPT plan works too on a Mac), and our team installs it with you. The Jony plugin works in Claude Code on a Mac or a Windows PC, and in Codex on a Mac. It also computes the figures on your computer. It needs Python 3.10 or later and your GitHub account added to our private store, and our team sets it up with you. The Jony connector works in Claude, Codex, GitHub Copilot, Cursor or any assistant that accepts a remote MCP server, with nothing to install. It serves the method and its roles and computes no figures. Microsoft 365 Copilot follows soon.
08We are invited. How do we start?
Two ways, with the account from your invitation. Install the desktop app, or connect Jony to Claude, Codex or Copilot in about two minutes.
09What does it cost?
Three things. Jony itself costs nothing during the beta, with no commitment beyond an NDA; pricing comes after, and you see it before anyone asks you to pay. Your firm brings its own model subscription (a Claude or ChatGPT plan, for example). And each person needs a Mac or a Windows PC with Microsoft Office for the Excel and PowerPoint deliverables. Each person reads up to 500 pages of the method a day; the count restarts at midnight UTC.
10What is it bad at?
Anything that needs a room. It will not read a management team, negotiate a peg or decide how hard to push an add-back. It also needs a real data room. Point it at forty scanned PDFs and it will tell you the room is incomplete.
Start on your next engagement.
Free during the beta. Ask for your invitation, then work in the app or in the assistant you already use.
- 1Request your invitation
- 2Download the app or connect your assistant
- 3Open an engagement folder
You are on the list
We open the beta to a few firms at a time, and we answer every request within two business days. A confirmation is on its way to your inbox. If you have a live deal coming up, reply to it and tell us: firms with a live engagement go first.