> 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/add-your-context/migrate.md).
# Import rules from CLAUDE.md and AGENTS.md
Import the rules in your CLAUDE.md, AGENTS.md, and rules folders into NeoHive, where every agent on the Hive can recall them.
Run one skill to turn the project rules in your context files, such as `CLAUDE.md`, into [Memories](/reference/glossary.md#memory). Every agent connected to your [Hive](/reference/glossary.md#hive) can then recall those Memories, with no file to commit, pull, or merge.
Your agent loads its whole context file, such as `CLAUDE.md` or `AGENTS.md`, on every task, even the parts that do not apply. After you import the rules and delete them from that file, your agent recalls only the rules each task needs.
## Run the migration
To start the migration, do the following:
1. Open your agent in the root folder of the project.
2. Run the skill, as shown for your agent in the following tabs.
{% tabs %}
{% tab title="Claude Code" %}
```
/neohive:migrate-memory
```
{% endtab %}
{% tab title="Cursor and Codex" %}
Ask your agent to run the `migrate-memory` skill.
{% endtab %}
{% endtabs %}
ChatGPT, Claude Desktop, and other MCP apps have no NeoHive plugin, so they cannot run the skill. Run the import once from Claude Code, Cursor, or Codex. Every agent connected to the Hive can then recall the Memories.
The skill looks for rules in the following places in the project, and uses the ones it finds:
* `CLAUDE.md`, `AGENTS.md`, and `GEMINI.md`
* The `.md` files in `.claude/rules/`. In Cursor and Codex, also the `.md` and `.mdc` files in `.cursor/rules/` and `.codex/rules/`.
* `CONVENTIONS*.md` and `CONTRIBUTING.md` files under `docs/`
The skill never imports the personal files in your home folder, such as `~/.claude/CLAUDE.md`, `~/.codex/AGENTS.md`, or `~/.cursor/rules/`.
## Review and approve the Memories
After you run the skill, the migration goes through the following steps:
{% stepper %}
{% step %}
## The skill lists the files
You see each project file that the skill reads. Files in your home folder show as skipped, because the skill reads them for context only.
{% endstep %}
{% step %}
## The skill sorts the rules
Each rule or decision becomes its own Memory with a [Memory type](/reference/memory-types.md), such as `directive`, `convention`, or `decision`. The skill leaves out setup steps, tables of contents, TODO lists, and personal preferences. The skill marks an entry that could be a personal rule or a team rule as ambiguous. The skill skips an ambiguous entry unless you re-classify the entry.
{% endstep %}
{% step %}
## You approve
If the skill asks which [Index](/reference/glossary.md#index) to use, select **Knowledge**. A preview then shows what the skill plans to store and what it skipped. The skill stores nothing until you approve. You can remove items, re-classify the ambiguous ones, or stop.
{% endstep %}
{% step %}
## The skill stores the Memories
If the Hive already holds a rule as a Memory, the skill skips that rule. The skill stores the rest in the [Knowledge Index](/reference/glossary.md#knowledge-index) and shows a summary.
{% endstep %}
{% endstepper %}
To approve each Memory one at a time, add `review=each` to the command: `/neohive:migrate-memory review=each`.
The skill does not change your original files. The skill also skips the NeoHive block between the `` and `` markers. The plugin writes that block to tell your agent how to use your Hive. The plugin updates the block through the markers, so keep the block and both markers in the file.
## After migrating
Your agent recalls migrated rules like any other Memory. A teammate on the same Hive gets the same rules without a copy of your context files.
The skill does not change your context files, so your agent still loads every rule in them on each task. To load only the rules that matter, remove the migrated rules from those files. Keep the rules the skill skipped, such as personal preferences, and the NeoHive block.
If a rule migrated incorrectly or becomes out of date, tell your agent. Your agent then deactivates the old Memory. For more information about what to say, see [Say "out of date" when a fact changes](/get-better-results/teach.md#say-out-of-date-when-a-fact-changes).
Long-form notes, such as an Obsidian vault, design docs, or runbooks, belong in a [Files Index](/reference/glossary.md#files-index). To upload long-form notes, see [Add documents and images](/add-your-context/documents.md).
---
# 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/add-your-context/migrate.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.
