For independent agencies & MGAs

Parse the loss run.
Score each claim. Skip the sheet.

Send submission documents in by email, a connected app, or your desktop. Lymnus reads every page in any language, extracts each field with a confidence score, cleans it, and delivers structured data to your systems or back by email — automatically.

No card needed SSO & SCIM built in 1% to climate

loss-run-38pg.pdf via email-in
vendor_name Harbor Mutual 99%
reference_id REF-2026-041 98%
total_amount 48,200.00 87%

↳ flagged for a 10-second human check

Bill created in QuickBooks Harbor Mutual · REF-2026-041 · 48,200.00
Would post to + 39 more

document in → clean entry out · no templates · no retyping

Your documents already contain the data. Someone just has to type it. Today that someone is you.

One PDF, dozens of claims to parse

A single carrier loss run holds claim after claim, and each field still gets typed into a sheet by hand.

Renewals hit all at once

Every account's history has to be rebuilt in the same window, field by field.

The first clean quote wins

Whoever has structured data ready quotes first — and usually takes the account.

How it works

A pipeline you set up once. Then it just runs.

01

Documents enter the pipeline on their own

Email them to a pipeline address, auto-import from a connected app, or upload from your desktop.

02

Every field extracted and scored

Scans, photos, handwriting, spreadsheet rows — parsed into your fields with a per-field confidence value.

03

Structured data routes where you quote

Passing records go to your agency management system, Sheets, or back by email.

The review queue

You review exceptions. Not everything.

Most documents flow straight through. When Lymnus is less than sure about a field — a smudged total, a handwritten date — that one document waits for a ten-second human check. Every correction makes your pipeline sharper.

A number that grows every week: the share of documents that never needed you at all.

The pipeline does more than read

Extraction is the first step — not the whole job.

Clean & merge, not just extract

Deduplicate, normalize dates and currencies, standardize vendor names, and merge many files into one clean dataset — as a step in the same pipeline.

Reads documents in 41 languages

A supplier invoice in German, a receipt in Japanese, a form in English — extracted into the same clean fields, in the language you work in.

Ask your documents anything

Once the data is in, the built-in assistant answers plain-English questions across everything your pipelines have processed — no formulas, no exports.

Reports that build themselves

Turn processed data into charts, summaries, and shareable reports — generated from your pipelines and exported wherever you work.

Built for teams

Share pipelines with your team, track every edit, fear no mistakes, and instantly revert incorrect data with a visual audit log, or add SSO and roles when you need them.

An API underneath it all

Everything the pipeline does is available over the API — push documents in, receive webhooks out, and build your own flows on top.

Ends up in your system, not ours.

QuickBooks Xero Google Sheets Excel Airtable Smartsheet Google Drive Dropbox OneDrive Notion PostgreSQL Snowflake Zapier Odoo Hubspot BigQuery PowerBi MySQL Salesforce Freshbooks MongoDB SharePoint API + webhooks
+ 39 more
SSO / SAML SCIM IP allowlists GDPR Enterprise →

Committed to Cleaner Compute.

Running software at scale consumes significant energy. That's why Lymnus commits 1% of revenue to verified CO₂ removal — offsetting the environmental cost of the infrastructure that powers your workflows, at no extra cost to you.

Your next 10 documents are on us.

No card needed