Know the covenant package before books close
Primary-market timetables do not leave room to read a 400-page indenture. ExactCov screens the package the hour it arrives, places it against precedent, and lets you ask the portfolio the questions that matter, each answer cited to its clause.
The document gets its one serious reading before close.
Every question afterwards goes back to the PDF. Capacity, portability, drop-down risk and consent thresholds are checked on the names someone remembers, and the summary written at close quietly stops being true after the first amendment. Under primary-market pressure, the summary of terms is read as if it were the document.
What a screen produces
In the hour the documents arrive, and kept current through every amendment.
Capacity and leakage
What the package allows the issuer to do to you.
- Day-one debt and lien capacity, ratio plus baskets
- RP capacity and unrestricted subsidiary transfers
- Value-leakage flags: drop-downs, uptiers, J.Crew and Serta patterns
Structure and events
What changes the picture.
- Portability and the leverage test it turns on
- Change of control, put rights and consent thresholds
- Cross-default reach and the order it fires in
Precedent
What the market gave on the last twelve deals.
- Each term placed against sector precedents
- Looser or tighter, by how much, with the clauses side by side
- Your own library, read once and kept
The covenant package nobody reads twice
One serious reading at close produces a filed PDF and a summary. Over the following years every capacity, headroom and consent question goes back to the PDF, while an amendment in year two leaves the summary out of date. A register is the reading you keep.
Read the pieceExtraction is where monitoring starts, not where it ends
Extraction produces a cited snapshot of the terms. Monitoring is the loop that runs against that snapshot for years: arrivals, tests, threshold changes, routing and stale detection. For the maintenance-covenanted names you hold, the same register runs the loop.
Read the pieceRight number, wrong clause
An analyst verifying a screen checks values. The failure that survives a spot check is the value that is right and the clause it points to that is not. That quadrant was 4.6 percent of fields before the validator and is 1.1 percent after it, which is why every figure in a screen carries a citation the analyst can open, and why the citation is checked before it is shown.
Measuring citation faithfulnessReading for buyside
Short pieces from the blog and the measurements behind them from Research.
The covenant package nobody reads twice
Documents get read hard once at close, then filed. Every question afterwards, capacity, headroom, permitted payments, sends someone back to a PDF. The problem, framed without naming a product.
Read Blog · 3 minHow much can they actually raise?
Incurrence capacity is the question that keeps returning and never has a stored answer. What a capacity table looks like, why it needs a history, and why it is a credit signal months ahead of the maintenance tests.
Read Blog · 3 min"What's market" is sitting in documents you have already read
Every desk reconstructs market terms from memory and a few recent deals. The distribution of every basket, cure and blocker across your own closed deals is a query once the corpus is in a register.
Read Blog · 3 minExtraction is where monitoring starts, not where it ends
LLM extraction from prospectuses and ISDA schedules works. What it does not give you is a register that stays live: a fresh NAV arriving, a test re-running, a threshold moving, a breach routed to the right desk.
Read Research · 3 minRight number, wrong clause: measuring citation faithfulness
Field accuracy is the metric everyone reports. It misses the quadrant that fails an audit: the value is correct and the clause it points to is not. That quadrant was 4.6 percent of our fields before the validator and is 1.1 today, inside a 97.4 percent first pass, and it is invisible to a spot check.
Read Research · 3 minHow deep do defined terms go? Definition chains in 300 agreements
A leverage covenant is one line. Resolving it means following Consolidated EBITDA into Consolidated Net Income and onward, a median of four hops and up to nine. Where the chains go, how often they leave the document, and what that does to extraction.
ReadThe parts investors use most
Everything on this page runs on the same register as the rest of ExactCov. These are the parts this team lives in.
Agreements Intelligence
Debt and lien capacity, restricted payments, portability, MFN and value-leakage flags on the day the documents arrive, every figure cited to its clause and every definition resolved to its leaf.
Agreements Intelligence PlatformExactCov on Claude and ChatGPT
Ask the register from the assistant your team already uses. Permissions mirror ExactCov, every answer carries its citation, and drafts wait for a person to send.
ExactCov on Claude and ChatGPT PlatformCommand Center
The whole book on one screen: what has arrived, what is stale, what is in breach and what is drifting, with headroom trends, scenario shocks and committee packs built from the register.
Command CenterRun it in parallel. Don't run a pilot.
The documents are loaded once. What already arrives by email is copied to a mailbox we watch. Nothing changes for the team, and after one cycle the two outputs are compared, disagreement by disagreement.
Screen a live deal with us
Bring an OM from a deal in market and the last three you bought in the sector. We show the package, the precedent comparison and the leakage flags, cited.