> For the complete documentation index, see [llms.txt](https://docs.gitloom.cloud/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.gitloom.cloud/readme.md).

# GitLoom

A git-backed memory engine for LLM agents, hosted or on your own machine.

GitLoom stores what an agent learns about a user as markdown in a git repository — one repository per namespace — and gives it back when a later question needs it. Git gives history, diffs, and blame on memories for free. Run it hosted over HTTPS, or entirely on your own machine with the CLI — same engine, same guarantees, no network on the local read path.

<a href="https://docs.gitloom.cloud/documentation/content/quickstart" class="button primary" data-icon="rocket-launch">Quickstart</a> <a href="https://docs.gitloom.cloud/api-reference/" class="button secondary" data-icon="terminal">API reference</a> <a href="https://app.gitloom.cloud/" class="button secondary" data-icon="key">Get an API key</a>

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><i class="fa-bolt" style="color:$primary;">:bolt:</i></td><td><strong>Quickstart</strong></td><td>From nothing to a retrieved memory in five requests.</td><td><a href="https://docs.gitloom.cloud/documentation/content/quickstart">https://docs.gitloom.cloud/documentation/content/quickstart</a></td></tr><tr><td><i class="fa-diagram-project" style="color:$primary;">:diagram-project:</i></td><td><strong>Core concepts</strong></td><td>Accounts, namespaces, memories, cues — and why the store is git.</td><td><a href="https://docs.gitloom.cloud/documentation/content/concepts">https://docs.gitloom.cloud/documentation/content/concepts</a></td></tr><tr><td><i class="fa-terminal" style="color:$primary;">:terminal:</i></td><td><strong>API reference</strong></td><td>Every route, request and response shape, every status code.</td><td><a href="https://docs.gitloom.cloud/api-reference/">https://docs.gitloom.cloud/api-reference/</a></td></tr><tr><td><i class="fa-code" style="color:$primary;">:code:</i></td><td><strong>TypeScript SDK</strong></td><td>Drop memory in beside the OpenAI or Anthropic SDK you already use.</td><td><a href="https://docs.gitloom.cloud/documentation/content/sdk">https://docs.gitloom.cloud/documentation/content/sdk</a></td></tr><tr><td><i class="fa-square-terminal" style="color:$primary;">:square-terminal:</i></td><td><strong>CLI &#x26; MCP</strong></td><td>Memory on your own machine, and an MCP server for your editor.</td><td><a href="https://docs.gitloom.cloud/documentation/content/cli">https://docs.gitloom.cloud/documentation/content/cli</a></td></tr><tr><td><i class="fa-tag" style="color:$primary;">:tag:</i></td><td><strong>Plans and pricing</strong></td><td>Free to evaluate, pay-as-you-go for exact usage, subscriptions for the rest.</td><td><a href="https://docs.gitloom.cloud/documentation/content/pricing">https://docs.gitloom.cloud/documentation/content/pricing</a></td></tr></tbody></table>

### Why git

{% columns %}
{% column width="50%" %}
**Correction without losing the record.** A fact that changes gets its file rewritten, not appended to — the memory always states what is true now, and the previous version is still in git's object store because that is what a commit is.

**Change detection is native.** Every node in the index is keyed by a git tree or blob hash, so re-indexing after a write is a tree diff, not a scan.
{% endcolumn %}

{% column width="50%" %}
**It is inspectable and portable.** A user's memory is a directory you can `git log`, diff, grep, and hand back to them — nothing is trapped in a schema nobody can read.

{% hint style="info" %}
Read the mechanism end to end in [How GitLoom works](https://docs.gitloom.cloud/documentation/content/architecture).
{% endhint %}
{% endcolumn %}
{% endcolumns %}

### Go deeper

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><i class="fa-gears" style="color:$primary;">:gears:</i></td><td><strong>How GitLoom works</strong></td><td>Git storage, incremental indexing, three-arm retrieval, fused.</td><td><a href="https://docs.gitloom.cloud/documentation/content/architecture">https://docs.gitloom.cloud/documentation/content/architecture</a></td></tr><tr><td><i class="fa-cloud" style="color:$primary;">:cloud:</i></td><td><strong>The platform</strong></td><td>The AWS architecture behind the hosted service, and what touches your data.</td><td><a href="https://docs.gitloom.cloud/documentation/content/platform">https://docs.gitloom.cloud/documentation/content/platform</a></td></tr><tr><td><i class="fa-chart-line" style="color:$primary;">:chart-line:</i></td><td><strong>Research and benchmarks</strong></td><td>Storage at scale, and the LongMemEval quality results end to end.</td><td><a href="https://docs.gitloom.cloud/documentation/content/benchmarks">https://docs.gitloom.cloud/documentation/content/benchmarks</a></td></tr></tbody></table>

Questions this page does not answer: <support@gitloom.cloud>.


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.gitloom.cloud/readme.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.