monkeSearch

Free plan LinuxMacWindows

What monkeSearch is

monkeSearch is an open-source desktop tool for finding local files with natural-language queries. Searches can combine file type, temporal context, and other keywords, and the maker says both approaches run fully offline. The main development branch uses a local language model to translate queries into macOS Spotlight arguments, relying on Spotlight’s index and the NSMetadataQuery API. This branch is currently macOS-only, searches indexed metadata rather than file contents, and requires macOS, Python 3.8 or later, llama-server, and a GGUF model; the stated default model is about 700 MB. A separate vectordb branch is described as cross-platform for macOS, Windows, and Linux, using vector databases instead of a live model at query time. Its Windows design crawls files with os.walk and stores retrieval data in ChromaDB. The project uses Apache-2.0 licensing and describes the Spotlight method as read-only with Spotlight access controls.

Who it suits

It suits people who want offline, natural-language searches over local files. macOS users can use the main branch, while users seeking Windows or Linux support should consider the vectordb branch.

What it does well

Runs searches fully offline
Natural-language queries combine multiple criteria
Open source under Apache-2.0
Spotlight approach is read-only
Vectordb branch is cross-platform

What to know first

Main branch currently macOS-only
Searches metadata, not file contents
Only Spotlight-indexed files are covered
Requires a local model setup

Limits

The main branch searches Spotlight-indexed files and metadata rather than file contents and is currently macOS-only. The repository also lists possible small-model misunderstandings and simple temporal expressions as limitations.

Getting started

Open the monkeSearch project website or repository
For the main branch, use macOS with Python 3.8 or later
Install or provide llama-server and a GGUF model, with the stated default model about 700 MB
Run natural-language queries using file type, temporal context, and other keywords
For cross-platform use, select the vectordb branch described for macOS, Windows, and Linux

Verdict

monkeSearch offers private, offline file discovery through natural-language queries. Platform choice, metadata-only search, and the setup requirements are central considerations.

Questions

Is monkeSearch free?

Yes. monkeSearch is free and open source under the Apache-2.0 license.

Which platforms does it support?

The main branch is currently macOS-only. The vectordb branch is described as supporting macOS, Windows, and Linux.

Does data leave the computer?

The maker describes both approaches as fully offline, with nothing leaving the machine.

What does the main branch require?

It requires macOS, Python 3.8 or later, llama-server, and a GGUF model. The stated default model is about 700 MB.

Can it search file contents?

The listed limitation is metadata-only search over Spotlight-indexed files, rather than content search.

How does the vectordb Windows approach work?

The described Windows architecture crawls the filesystem with os.walk and uses ChromaDB for vector storage and retrieval.

Facts from monkeSearch's own pages

Agent use: The maker says the LLM-based implementation can serve as a natural-language file-discovery bridge for AI agents and orchestration systems. · monkesearch.github.io, 5 Oct 2026
Benchmark: The maker's benchmark page says its query-parsing evaluation covers 80 queries across file type, temporal awareness, combined queries, and noise resistance. · monkesearch.github.io, 11 Oct 2026
Current platform status: The repository describes the main branch as currently macOS-only and says cross-platform support is in the vectordb branch. · github.com, 11 Oct 2026
Data handling: The maker says nothing leaves the machine; the Spotlight approach is read-only and uses Spotlight's access controls. · monkesearch.github.io, 11 Oct 2026
Integrations: The main/dev approach uses macOS Spotlight and is described as a natural-language filesystem bridge for AI pipelines and agents. · monkesearch.github.io, 11 Oct 2026
License: The repository lists the project under the Apache-2.0 license. · github.com, 11 Oct 2026
License and support: The project is open source under the Apache-2.0 license, and its site asks users to support it by starring the repository. · github.com, 5 Oct 2026
Limitations: The repository says the main implementation searches Spotlight-indexed files, searches metadata rather than file contents, and supports only simple time expressions. · github.com, 11 Oct 2026
Local processing: The project says both its LLM and vector database approaches run fully offline. · monkesearch.github.io, 11 Oct 2026
macOS approach: The main/dev branch uses a local LLM to turn queries into macOS Spotlight search arguments. · monkesearch.github.io, 11 Oct 2026
macOS integration: The main/dev branch uses Spotlight’s existing index and the NSMetadataQuery API for file discovery. · monkesearch.github.io, 5 Oct 2026
Performance tradeoff: On macOS, the maker says disabling LEANN recompute makes searches faster at the cost of disk space; recompute is enabled by default. · monkesearch.github.io, 5 Oct 2026
Platform architecture: The maker describes vectordb support for macOS, Windows, and Linux; the repository says the main branch is currently macOS-only. · github.com, 5 Oct 2026
Product: monkeSearch is an open-source, offline-first desktop tool for searching local files with natural-language queries. · monkesearch.github.io, 5 Oct 2026
Purpose: monkeSearch is an offline-first desktop tool for searching local files with natural language. · monkesearch.github.io, 11 Oct 2026
Requirements: The repository lists macOS, Python 3.8+, llama-server from llama.cpp, and a GGUF model as requirements. · github.com, 11 Oct 2026
Search inputs: The maker describes queries as combining file type, temporal context, and other keywords. · monkesearch.github.io, 5 Oct 2026
Search queries: Queries can combine file type, temporal context, and other keywords. · monkesearch.github.io, 11 Oct 2026
Security: The maker describes the Spotlight-based approach as read-only and scoped through Spotlight’s access controls. · monkesearch.github.io, 5 Oct 2026
Support: The project page asks users who find the open-source project useful to star its GitHub repository. · monkesearch.github.io, 11 Oct 2026
Vector database approach: The vectordb branch uses vector databases instead of a live LLM at query time and is described as cross-platform. · monkesearch.github.io, 11 Oct 2026
Windows integration: The vectordb Windows architecture uses an os.walk filesystem crawl and ChromaDB for vector storage and retrieval. · monkesearch.github.io, 5 Oct 2026

Where monkeSearch stands on Freedom251

Desktop Search Software#21 of 57