LatestAll news

Home / Use cases / PCT-013

Underwrite · PCT-013 · Sourced card

Track every amendment and waiver

Reads every amendment and waiver across a book and logs what changed, in which clause, and whose consent it needed.

Worked example: Amendment No. 1 adds one new defined term, HSBC ME Obligations (capped at $75m), and it now appears in the maturity date, the payment waterfall and the audit-opinion carve-out. See the walk-through ↓

Special sits / distressedDirect lendingPublic + sample docsVendor + in-house
INPUTAmendments and waiversAI STEPSRead all amendments · Build a datedchange log · Draft what we give upOUTPUTDated change log with what each changegives upHUMAN CHECKHow to vote and what to ask for inreturn

Today vs automated

TODAY
  1. Read each amendmentAssociate · manual
  2. Keep a change log by handAssociate · manual
  3. Compare the new consentVP / counsel · manual
  4. Email / meeting hand-offAssociate → VP / counsel
AUTOMATED
  1. An amendment or waiver arrives, or is filed on EDGARTrigger: starts on its own
  2. Read all amendmentsAutomated · Document parser
  3. Build a dated change logAutomated · LLM
  4. Draft what we give upAI draft, human review · Model template
  5. How to vote and what to ask for in returnApprove / override · VP / counsel

AutomatedAI draft, human reviewHuman onlyTools: Data room, Document parser, LLM, Model template

Estimated valueestimate

50–920 analyst hours a year
$8k–$231k of time, vs an assumed $20k–$100k tool cost

Net: −$92k to $211k a year net of tool cost; at the low end of the assumptions it does not cover the tool cost.

Manual2–4 h per amendmentWith AI0.3–1 h per amendment, review included
Assumptions
  • 2–4 analyst hours per amendment today; 0.3–1 hours with an agent, including review
  • 50–250 amendments a year for one team
  • $150–250 per hour loaded cost
  • Tool or build cost $20,000–100,000 a year (both)

Estimate, based on the assumptions shown; your numbers will differ. How we estimate

What else it can do for the team

  • Map each change to the consent class it needed
  • Update the covenant and basket table
  • Flag soft credit events for the watchlist
  • Draft the IC note on the amendment
  • Keep a fee and pricing history per facility
Worked walk-throughKey catch: Amendment No. 1 adds one new defined term, HSBC ME Obligations (capped at $75m), and it now appears in the maturity date, the payment waterfall and the audit-opinion carve-out.

1. The input

A public credit-agreement amendment: Amendment No. 1 dated September 8, 2026 to IBEX’s U.S. credit agreement (HSBC Bank USA as agent), filed as Exhibit 10.1 to an 8-K on September 10, 2026. Source: Exhibit 10.1 on SEC EDGAR · the 8-K.

2. What the AI produces

What changedNew text saysType
“Maturity Date” (§1.1)Restated: earlier of October 22, 2029 and termination or maturity of any HSBC ME ObligationsExtension, now tied to a second facility
“Applicable Commitment Fee Rate” (§1.1)Restated: 0.30% per annumPricing
“HSBC ME Obligations” (new, §1.1)Obligations under the affiliate’s HSBC Bank Middle East facilities, capped at $75,000,000New cross-facility link
§5.1(a)(i)(x) audit opinionGoing-concern carve-out now also covers impending maturity or covenant breach under the HSBC ME loan documentsReporting covenant loosened
§7.2 waterfall, clause SixthRestated to include HSBC ME Obligations that became due earlyPayment priority change
Exhibit GForm of compliance certificate restated (Annex I)Reporting form
§3(d) conditionAmendment fee of 0.20% of revolving commitments ($50,000)Fee

Values read from the filed exhibit on 2026-10-18. A real log would add the before text from the original agreement for each row.

3. What a person still checks

  1. Pull the original agreement and confirm the before text for each restated definition.
  2. Check whether the change needed Required Lender or all-lender consent under the amendments section.
  3. Trace knock-on effects: a new defined term (here, HSBC ME Obligations) can enter the waterfall, cross-default and Material Indebtedness at once.
  4. Decide what the change means for your position; the log records it, it does not judge it.

4. Where it breaks

  • Amendments that restate whole definitions hide the actual delta; without the prior text the tool reports what is new, not what changed.
  • Side letters, fee letters and waivers granted by email never reach EDGAR or the data room.
  • Conformed copies and serial amendments get out of order if dates are not checked.
How you get thereFour steps from prompting to an agentic workflow; most teams start at step 4.
1 · PromptingAn analyst asks a chat tool about one amendment.
Needs: an approved AI tool, a confidentiality policy.
Risk: unchecked answers, data pasted outside policy.
2 · WorkflowA fixed pipeline extracts the same fields into your template every time.
Needs: field definitions, a template, document parsing.
Risk: breaks silently when formats change.
3 · AgentIt watches the inbox or data room, runs the checks and drafts the output.
Needs: inbox/doc-store access, system write-back, logging.
Risk: acting on a misread; needs review gates.
The automated lane above
4 · Agentic workflowSeveral agents work across the deal or book; people approve exceptions.
Needs: shared data model, permissions, audit trail, exception queue.
Risk: errors compound across steps; needs monitoring.
Typical starting point
Solution optionsBuy (22 vendors listed), build or hybrid, with the trade-offs of each.
Build
An LLM API, a document parser, a vector or document store, your data warehouse, and a review queue.
Fits your definitions exactly; you own maintenance and controls.
Hybrid
Vendor extraction feeding your own checks, templates and systems.
Common middle path; two things to integrate and govern.

Listing is not endorsement.

What changes for the team2–4 h → 0.3–1 h per amendment (estimate); coverage goes from the sections someone had time to read to every clause and schedule checked against the term sheet.
ROLESAssociates move from reading and retyping to checking cited exceptions.
CYCLE TIME2–4 h → 0.3–1 h per amendment (estimate)
COVERAGEFrom the sections someone had time to read to every clause and schedule checked against the term sheet.
CONTROLSEvery output cites its source; approvals and overrides are logged.
How it works

Full title: Soft credit events, read overnight: every amendment, waiver, PIK toggle and extension in your book

“Moody's says counting amendments and PIK lifts the default rate from 4.0% to 5.8%. Here's how you count them in your own book.”

This card now includes former PCT-031 (Monitor a name like a terminal: every filing, 8-K, and mark move in one feed).

Scenario

SAMPLE: A fund's book has 60 credit agreements and about 200 amendments. A new consent request arrives on fictional 'Redwater Foods' at 6pm, due in 48 hours.

What the human still decides

How to vote and what to ask for in return.

Why it matters

Moody's found that counting PIK amendments and maturity extensions among distressed borrowers lifts the private credit default rate from 4.0% to 5.8%. Amendment voting is also going digital (Versana, Oct 2026), so requests arrive faster.

What the record shows

Checked 2026-10-15.

Moody's: including PIK amendments and maturity extensions among distressed borrowers raises the 12-month default rate on 607 credit-estimate borrowers from 4.0% to 5.8%. That gap is what an amendment tracker measures. Plumbing is arriving: Versana's digital loan voting went live with five agent banks, citing coverage of 11,000+ facilities and $4.8tn of commitments.

Try it: prompt and data
Here is a credit agreement and its amendments in date order. Build a change log table: date, section, what changed, before → after. Then compare the attached consent request with the current terms and list what lenders would give up.

Data

Public filings plus clearly labeled SAMPLE documents. Never paste confidential deal data into a consumer AI tool.

Who uses it, vendors and build
Segments
Special sits / distressed, Direct lending
Who uses it
Special sits analyst, Portfolio manager, Counsel
Build
Vendor + in-house
In-house
Document diffing + LLM summarizer + a strict table schema. Estimate: 2–3 weeks.
Last verified
2026-10-15

Vendors that claim this capability

Tags reflect each vendor's own public claims at the linked page (checked 2026-10-15). Not tested, ranked or endorsed.

Sources and related
In the news.

Part of the deal lifecycle: stage 4 of 7, Documents.

Next step

Get new walk-throughs

Subscribe to the newsletter

Not investment advice. Names marked SAMPLE are fictional. Found an error? See the corrections log.