> For the complete documentation index, see [llms.txt](https://docs.neohive.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.neohive.ai/welcome.md). # Welcome NeoHive gives your coding agents your codebase and your team's knowledge, from a server you run yourself. **Coding agents that share what your whole team knows.** NeoHive is a self-hosted context layer: one memory that your team's coding agents share. Connect your repositories, documents, and team knowledge once, and every agent on your team reads from that memory. That includes Claude Code, Cursor, Codex, and any other agent that supports the [Model Context Protocol (MCP)](/reference/glossary.md#mcp). When one teammate corrects an agent, every agent on the team gets the correction. NeoHive updates the memory as your repositories change. Each note your team saves also improves the answers that every agent gets. That memory lives on machines that you control, because NeoHive runs inside your network. ## Start here


Quickstart	Install NeoHive and get your first answer from your own code.	Quickstart
Your first session	Ask about your code, teach one convention, and see the convention come back next session.	Your first session
How NeoHive works	Learn about Hives, Indexes, and how your agent reaches them.	How NeoHive works
Add a repository	Connect your repositories, such as GitHub or GitLab repositories.	Add a repository
Import rules from CLAUDE.md and AGENTS.md	Bring your existing context files into NeoHive.	Import rules from CLAUDE.md and AGENTS.md
Teach your agent as you work	Correct your agent once, and every teammate's agent remembers it.	Teach your agent as you work

## Why it exists Coding agents fall short on a team in the following ways: * **Every developer teaches their own agent.** A correction that one person makes in Claude Code doesn't reach a teammate's Cursor. * **Every agent starts from zero.** Your agent doesn't know your codebase, your conventions, or the library that you moved off last quarter. * **Context files don't scale.** A `CLAUDE.md` file can grow to thousands of lines, and the agent loads the whole file, even when the task needs only a small part of it. * **Reasoning is the first thing lost.** Teams rarely write down why they made a decision, but that reason is what the next agent needs most. NeoHive gives your team one persistent memory that every agent reads from and every teammate adds to. That memory lives in a [Hive](/reference/glossary.md#hive), your team's workspace. A Hive holds [Indexes](/reference/glossary.md#index), which store your code, your documents, and the [Memories](/reference/glossary.md#memory) that your team saves. For more information about the problem and how NeoHive solves it, see [What is NeoHive?](/get-started/what-is-neohive.md). ## What NeoHive does


Knows your codebase	NeoHive indexes your repositories, so your agent finds code by what it does, not by guessing file names.	How retrieval works
Reads your documents	NeoHive indexes the specs, runbooks, and PDFs you upload to a Files Index, so your agent recalls them like code.	Add documents and images
Stays up to date	NeoHive syncs your repositories on a schedule. Each sync indexes only the files that changed.	Keep a repository up to date
Keeps your team's knowledge	NeoHive stores conventions, decisions, and debugging notes as Memories. Everyone's agent can recall what one person teaches.	Capture team knowledge
Works with every agent	NeoHive connects to Claude Code, Cursor, and Codex through plugins, and to ChatGPT, Claude Desktop, or any other MCP app directly.	Connect your agent
Runs on your machine	NeoHive keeps your code, documents, Memories, and recall queries on the machine it runs on. NeoHive doesn't send that content to an outside service.	What stays on your machine

## Built for teams A Hive is shared, so what one person teaches their agent helps the whole team: * **One Hive, every agent.** Your teammates' agents read from the same Hive, whichever tool each teammate uses. * **Teach once.** When you correct your agent or explain a convention, the agent saves that note as a Memory in your Hive. Every teammate's agent then recalls that Memory. * **Notes outlast team changes.** The note stays in the Hive, so someone who joins next month has it on their first day. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.neohive.ai/welcome.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.