Sifter MCP server

Sifter is a tool that reads your documents, such as invoices, contracts or CVs, and pulls out the important details into a neat table. Once that table exists, you can ask questions about it in plain English and get exact answers. It is handy if you deal with lots of similar documents and want totals, counts or filtered lists instead of reading them one by one.

What this MCP server does

You upload your documents to Sifter and describe what to pull out, for example client name, date and total amount. Sifter reads each document and turns those details into typed records, like rows in a spreadsheet. When you ask your AI a question, the AI sends it to this helper, the helper asks Sifter, and Sifter runs a real database query over those records. You get back exact rows, counts and sums, not a vague text summary. The whole loop happens inside your chat.

Click to zoom

What you can do with it

• Ask for totals and counts across many documents at once
• Filter records by date, name, amount or any field you extracted
• List documents that match a condition, like contracts expiring soon
• Create new sifts and upload documents through the AI
• Look up a single record by its details
• Build simple dashboards from a plain language description
• Inspect the query Sifter ran, so you can see how the answer was produced

Try asking your AI

• “What's the total unpaid across all invoices from last quarter?”
• “Show me all contracts expiring in the next 90 days.”
• “Which candidates have Python and more than 5 years experience?”
• “List every invoice from Acme Corp above 1000.”

What it gives back to you

You get back structured answers: exact numbers, filtered lists of records, or a small table of rows with the fields you asked about. For example, a total amount, a count, or a list of client names with dates. If the AI created or changed something, like a new sift or an upload, it tells you what it did. The answers are reproducible, because they come from a database query rather than a guess.

Before you start

Who it's for

Anyone who handles lots of similar documents, like finance teams, recruiters, legal and operations staff, and wants quick exact answers without building a database.