Skip to content
MCP server

Lochless

Search a document set and return every relevant match with quotes

Install
Pricing
Freemium
Platforms
API, Web, self-hosted
Runs
Remotely
Version
0.1.2
Listing updated
1 day ago

About Lochless

Lochless is a freemium MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Lochless takes a different route from top-k retrieval. It reads the documents in scope, up to a spending limit you set, and returns each one it judges relevant as a quote with your own document id and metadata. There is no chunking code, embedding pipeline or vector database to run.

The MCP server exposes search_documents, browse_documents, read_document and ask_documents, with OAuth sign-in or the same API key as the REST calls (ingest, context, ask). A free demo endpoint serves 20 public US founding documents so you can try it without an account. Browsing and reading are free; searching and asking are billed by text scanned.

Published prices include $0.005 per page indexed and $0.05 per /ask answer, and new workspaces get $5 of credit without a card. The vendor's own benchmark reports slightly lower F1 than three data platforms on yes/no classification, at a lower price per million tokens, and says so openly.

Install

Run in a terminal

claude mcp add --transport http lochless https://api.lochless.io/mcp

Add to Settings → Connectors → Add custom connector

https://api.lochless.io/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "lochless": {
      "url": "https://api.lochless.io/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "lochless": {
      "type": "http",
      "url": "https://api.lochless.io/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "lochless": {
      "serverUrl": "https://api.lochless.io/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "lochless": {
      "type": "http",
      "url": "https://api.lochless.io/mcp"
    }
  }
}

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

search_documents
Searches documents and returns relevant quotes.
browse_documents
Browses indexed documents.
read_document
Reads a document.
ask_documents
Answers a question with citations.

What it does

Exhaustive retrieval
Returns every relevant document in scope rather than a fixed top results list.
Cited answers
The ask call writes a short answer that cites its evidence inline.
Read limits and dry runs
A dollar read limit and a free dry run show coverage and cost first.
Metadata and date filters
Narrow queries by time range or exact-match metadata.
Self-hosting option
Teams can run it in their own AWS or GCP account on an annual plan.

Good for

  • Find every support ticket that mentions refunds in a month
  • Pull all clauses on a topic from a set of contracts
  • Give an agent a searchable archive without running a vector database
  • Ask a cited question across transcripts or logs

Pricing

Freemium

$5 free credit; $0.005 per page indexed; $0.05 per answer plus tokens scanned

Prices change. Check lochless.io before you buy.

Lochless FAQ

How is it priced?

Indexing is $0.005 per page, /ask is $0.05 per answer, and queries pay for text scanned.

Can I try it first?

Yes, new workspaces get $5 free credit with no card, and a demo MCP endpoint uses public documents.

Which MCP tools exist?

search_documents, browse_documents, read_document and ask_documents.

Is Lochless free?

Lochless has a free plan, with paid plans for more usage or features.

How do I add Lochless to Claude Code?

Run this command in your terminal: claude mcp add --transport http lochless https://api.lochless.io/mcp

What are the best alternatives to Lochless?

The closest alternatives to Lochless are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

Alternatives to Lochless

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  4. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  5. note1

    MCP server

    MCP server giving AI assistants search and summaries across your note1 meetings

    • 4 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Productivity