Truth in Lending (Regulation Z) Annual Threshold Adjustments (Credit Cards, HOEPA, and Qualified Mortgages)
Raises the HOEPA points-and-fees dollar trigger from $1,348 to $1,380 and the total-loan-amount threshold to $27,592, effective 2026-01-01.
Every week, the CFPB, FHFA, HUD, the GSEs, and 54 state regulators publish dozens of pieces of guidance. Final rules, proposed rules, bulletins, mortgagee letters, circulars, procedure notices. Most compliance teams keep up via a mix of manual RSS monitoring, lawyer alerts, and luck. Nothing aggregates it. Nothing tells you which pieces are relevant to your operation.
Missed rules become enforcement risk. A consent order, an exam finding, a loan buyback — these are the downstream costs of reactive compliance. And reactive compliance is how the industry operates by default, because keeping up manually is genuinely hard.
Existing tools index the firehose but stop there. They give you a list of documents. They don't tell you which rule actually affects your policies, and they don't draft what you'd need to change. That's the gap Reglith fills.
We watch every regulator that matters for residential mortgage compliance, summarize what changed, and map it to your own policies. The work that used to take a compliance officer half a Monday — triaging the week's guidance, checking what applies, noting what needs a policy update — is something the product does in the background.
We sell to small and mid-size lenders. The plan is $79 per month, no tiers, no haggle, because we don't want to spend any of our time selling. You either find it useful at that price or you don't. No enterprise contract, no implementation project, no dedicated success manager.
Impact finding (illustrative)
HOEPA threshold change affects your High-Cost Mortgage Policy
CFPB · Final rule · 12 CFR 1026.32(a)(1)(ii)(B)
§3.2 uses the 2025 points-and-fees trigger of $1,348; the rule sets $1,380 effective 2026-01-01.
Example: the rule is real (Federal Register 2025-22773, Dec 15, 2025); the policy and the finding are a sample.
Real data, first.
Every rule we ingest is linked back to its source document. The AI summaries are not the source of truth — the rule is. We log the prompt, model, and raw output for every summary so we can audit accuracy when a question comes up. The original document is one click away on every item in the feed.
Transparent ingest.
The full list of sources we cover is public at /sources. When ingest goes stale — if a regulator changes their page structure and our parser starts returning nothing — we tell you. We run a daily health check and flag degraded sources rather than quietly serving outdated data.
Indie-built, founder-supported.
If something is wrong, you can email the person who built it. No tier 1 / 2 / 3 support queue, no SLA that only applies at the enterprise tier. This is a focused tool built by a single developer who works in the problem space. The feedback loop is short by design.