Open-source Mindbreeze alternatives

Onyx is an open-source AI platform that acts as a knowledge and context layer for teams and their AI agents. It connects to more than 50 applications, pulls in documents along with their metadata and permissions, and builds an internal representation so that LLMs can answer questions grounded in what your company actually knows.

The project argues that this indexed approach gives more reliable, lower-latency and cheaper context than approaches that have an agent search many tools iteratively through MCP. Features include agentic RAG on a hybrid index, a Deep Research mode for multi-step reports, custom agents with their own knowledge, instructions and actions, and live web search through providers such as Serper, Google PSE, Brave and SearXNG. Sandboxes and skills extend what models can do, and the chat interface works with every major LLM.

Onyx emphasizes data sovereignty through flexible self-hosted deployments, with a single-command deploy option. The code is written in Python and Next.js, and the license is listed as 'Other' on GitHub, so check the terms for the enterprise features. It suits organizations that want a private, ChatGPT-style assistant connected to internal documents.

SWIRL is an AI search and retrieval-augmented generation tool that works without moving your data. Instead of copying content into a vector database, it queries your sources live with each user's own permissions, re-ranks the results, and can generate an answer with citations using the language model of your choice. The Community edition in this repository is released under the Apache-2.0 license.

Connectors reach many business applications, and the project's tagline refers to more than 100 apps. The approach avoids building ETL pipelines, standing up a vector store, or maintaining a second copy of data that needs security and audit controls, since permissions are enforced at the source. The stack is Python and Django, and a Docker quick start gets an instance running in about two minutes.

A separate SWIRL Enterprise edition adds a three-pass reranker, canonical answers, an MCP server for agents, and managed support. The Community edition is free to self-host. It suits IT and data teams wanting enterprise search across many systems while keeping data in place.

Fess is an open-source enterprise search server that you install and run on any platform with a Java runtime. It is built on OpenSearch but does not require prior OpenSearch knowledge, because everything is configured in a browser-based administration interface. It can also serve as a free site search for your own website.

A built-in crawler collects documents from web sites, file systems and data stores such as databases, CSV files, cloud storage and SaaS sources, and handles many formats including Microsoft Office files, PDF and archives. Search supports faceting, sorting and suggestions, results can be filtered by role and permission, and single sign-on works with LDAP, OpenID Connect, SAML, SPNEGO and Microsoft Entra ID. It offers a REST API, text analysis for over 20 languages, plugins, and AI features for RAG and semantic search.

Fess is written in Java and licensed under Apache-2.0. It needs Java 21 or newer for the ZIP, RPM and DEB packages, and Docker images bundle OpenSearch. It suits organizations that want a private, self-managed alternative to Elasticsearch-based setups or Google Site Search.

Xyne is an AI-first search and answer engine for work, offered as an open-source alternative to enterprise search products such as Glean, and to Gemini and Microsoft Copilot in the same space. It connects to the applications where work information is scattered, indexes the data securely and maps relationships between items, so users can search and ask questions in a Google-plus-ChatGPT style with sources.

At the time of the README, it covered the whole Google Workspace, including Drive files, PDFs, documents, sheets and slides as well as calendar, contacts and Gmail with attachments, with Slack and Atlassian integrations planned next. It can be self-hosted on a laptop, on-premises or in the cloud, is model-agnostic so it can use any LLM or cloud provider, including a local model through Ollama, and emphasizes private, secure use without training on your data.

Important: the README states that this repository is no longer maintained and has moved to github.com/juspay/xyne-spaces. The code here is written in TypeScript with Bun and licensed under Apache-2.0; new users should evaluate the new repository.

Open Semantic Search is a research tool aimed at journalists, researchers and analysts who need to search, browse, analyze and explore large collections of documents, going well beyond simple keyword search. It combines a semantic search engine with a full text mining and text analytics platform.

It includes an ETL framework for processing documents, covering crawling, text extraction, text analysis, named entity recognition for persons, organizations and locations, and OCR for images and embedded images inside PDFs. On top of that data pipeline, it provides metadata management through thesauri and ontologies using the SKOS standard, and search user interfaces and applications supporting full-text search, faceted search, exploratory search and knowledge graph search. Its topics specifically mention OSINT and investigative journalism as use cases.

Open Semantic Search can be built as a Debian or Ubuntu package, as a desktop search appliance for VirtualBox, or as a set of Docker images using the project's docker-compose configuration, which starts all the needed services together. It is written in Shell and Python, documented with MkDocs, and released under the GPL-3.0 license.