Read the data room. Trace every number.
Infratex turns CIMs, financial statements, credit agreements and board packs into structured context your diligence agents can query, with every figure linked to its page.
Every number in a data room has a page behind it. Infratex reads the CIM, the management accounts and the credit agreements, keeps every table intact, and gives your agents figures they can trace back to where they were written.
One question across the portfolio
An illustrative portfolio of 14 companies with their credit agreements and latest quarterly reports. The answer cites the covenant and the reported figure it is checked against.
Question
Is any portfolio company below its DSCR covenant this quarter?
Answer
One is. Acme Robotics reported a debt service coverage ratio of 1.12x for Q31, below the 1.25x minimum in its credit agreement2. The other thirteen are above their thresholds, including Nexus Health at 1.62x against a stricter 1.50x covenant.
Covenant compliance
| Q2 2026 | Q3 2026 | |
|---|---|---|
| EBITDA (€k) | 4,812 | 4,105 |
| Debt service (€k) | 3,540 | 3,665 |
| Debt service coverage | 1.36x | 1.12x |
22. Financial covenants
22.1 Debt service cover. The Borrower shall ensure that the Debt Service Coverage Ratio for each Relevant Period is not less than 1.25:1, tested on each Quarter Date by reference to the most recent Compliance Certificate.
How Infratex reads a data room
Parse
Tables that keep their shape
Financial statements, footnotes and long tables come back as Markdown, row for row, with a marker for every page. Methods that emit OTSL also keep merged cells and nested headers. In our run of the PulseBench-Tab table benchmark, Infratex Standard ranked first of eight systems.
<!-- page:14 --> ## Covenant compliance | | Q2 2026 | Q3 2026 | |----------------|--------:|--------:| | EBITDA (€k) | 4,812 | 4,105 | | Debt service | 3,540 | 3,665 | | DSCR | 1.36x | 1.12x |Index
One index over every document
A hybrid index spans the CIM, the accounts, the contracts and the board packs, so a question can draw on the cohort table as well as the commentary around it.
query "debt service coverage" 1 Acme_Robotics_Q3_2026_report.pdf page 14, Covenant compliance 2 Acme_Robotics_credit_agreement.pdf page 84, clause 22.1Answer
Figures with their sources
Ask in plain language, or define a schema and pull the same KPIs and terms from every company as JSON, with the page behind each value.
{ "company": "Acme Robotics", "period": "Q3 2026", "dscr": 1.12, "dscr_covenant_min": 1.25 }
The same terms from every deal
Define the fields once as a template and run it on every term sheet and side letter. Each value comes with its evidence, and each run exports to CSV or XLSX.
| Company | Round | Liquidation preference | Anti-dilution | Pro rata | Board |
|---|---|---|---|---|---|
| Acme Robotics | Series Bp. 1 | 1× non-participatingp. 3 | Broad-based weighted averagep. 5 | Major investorsp. 6 | 1 of 5 seatsp. 7 |
| Nexus Health | Series Ap. 1 | 1× non-participatingp. 2 | Broad-based weighted averagep. 4 | All investorsp. 5 | 1 of 3 seatsp. 6 |
| Tallow Systems | Series Bp. 1 | 1.5× participating, 3× cap (departs from the standard terms)p. 3 | Full ratchet (departs from the standard terms)p. 4 | Major investorsp. 5 | 2 of 5 seatsp. 6 |
| Brightwater Bio | Seed extensionp. 1 | 1× non-participatingp. 2 | Broad-based weighted averagep. 3 | Nonep. 3 | Observerp. 4 |
Where investment teams use it
- Due diligence
- Read CIMs, management accounts and customer contracts across the whole data room.
- Portfolio monitoring
- Pull the same KPIs and covenant figures from every company’s quarterly reporting.
- Deal terms
- Compare economics and control terms across term sheets, side letters and rounds.
- Investment memos
- Draft from figures that each come with the page they were read from.
Runs where your documents live
Managed cloud
Use the hosted API with a key. Nothing to install or run.
Your private cloud
Run Infratex inside your own VPC, next to the documents it reads.
On-premises
An isolated container deployment for teams whose documents cannot leave the building.
See it on your own documents
Tell us where your document pipeline breaks and upload a few samples. We'll run them through Infratex and walk through the structured output with you on the call.