About Localsynapse
Localsynapse is a Windows app for searching local files and email, aimed at people who need keyword or meaning-based discovery and AI access to documents. Search, indexing, and embeddings run on the machine, with index access through a built-in MCP server.
Key features
- 01Hybrid full-text and semantic search
- 02Support for many document formats
- 03Preserve Excel spreadsheet structure
- 04Search local email
- 05Run indexing and embeddings locally
- 06MCP access with source attribution
Use cases
- 01Find files by keyword or meaning
- 02Search email by sender, subject, or body
- 03Let an AI assistant read documents and spreadsheets
- 04Retrieve local content with source attribution
MCP support
Its built-in MCP server lets AI assistants search the local index and read structured spreadsheet views with source attribution.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMELocalSynapse
AI-powered local search for your files and email on Windows. Find anything by keyword or by meaning — and let your AI assistant read your documents and spreadsheets — all on your own machine.
Features
- Hybrid search — full-text (BM25) and semantic (local AI embeddings) together, with first-class Korean/English handling.
- Reads your formats — PDF, Word, PowerPoint, Hangul (HWP), CSV, Markdown, plain text, and more.
- Faithful spreadsheets — Excel files are read with their structure intact: cell coordinates, merged ranges, and correct types (dates stay dates, numbers stay numbers). Large workbooks parse quickly, up to 25 MB.
- Local email search — your mail in
.eml,.msg,.mbox, and exported mailbox JSON is searchable alongside your documents — by sender, subject, and body. - Works with your AI — a built-in MCP server lets assistants like Claude search your index and read a faithful, structured view of your spreadsheets — locally, with source attribution.
- Local-first & private — search, indexing, and AI embedding run on your machine, and your files and their content never leave your device.
Download
Get the latest installer from the Releases page.
Project site: localsynapse.com
What's in this repository
This repository hosts release artifacts only — Windows installers and their SHA-256 checksums on the Releases page. Active development happens in a private repository; the source tree is no longer published here from v2.15.0 onward.
Historical source from v1.0.0 through v2.14.0 remains in the repository's git history per the Apache License 2.0 under which those versions were distributed. Anyone who needs a snapshot of those versions can recover them from the tag history or from public forks.
License
Apache License 2.0 — see LICENSE.
The license covers the distributed binaries and all source code that was published in this repository through v2.14.0. v2.15.0 and later are distributed as installer binaries only.
Issues and questions
Use the Issues tab to report bugs, request features, or ask questions about LocalSynapse.
— Ryan
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.