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

# Libraries

Libraries are how you give the Notara agent knowledge about your organization. A library is a collection of context documents — your company handbook, API documentation, runbooks, competitor research, meeting notes, and anything else your team wants the agent to reference.

## What Is a Library?

A **library** is a named collection of documents, scoped to a **group**. When the agent responds to a request in a channel that has a library assigned, it draws on those documents as part of its context.

Libraries use a **manifest** — an auto-generated index document the agent reads first. The manifest lists every document in the collection by title, type, and a one-line description. This means the agent always knows what's available, even if it can't fit every document into its context window.

## How the Agent Uses Libraries

Each document in a library has a **load policy** that determines when it's included:

| Policy      | Behavior                                                  |
| ----------- | --------------------------------------------------------- |
| `spine`     | Always injected into every agent turn. Budget-capped.     |
| `on_demand` | Retrieved via semantic search when the query is relevant. |
| `never`     | Never shown to the agent. Human reference only.           |

The agent's context budget ensures it never exceeds the model's token limit. Spine documents fill the budget first; the manifest then lists all `on_demand` documents so the agent can request specific ones if needed.

→ [Detailed explanation of load policies](/notara-docs/libraries/load-policies.md)

## Groups and Library Scoping

Libraries are scoped to **groups** — sub-teams within your workspace. This means:

* Your engineering group can have a library with API docs, system design docs, and postmortem runbooks
* Your marketing group can have a library with brand guidelines, competitor research, and campaign templates
* A shared "company" library can be assigned to multiple groups

Each group only sees the libraries assigned to it. Documents in a group's library don't leak to another group's agent sessions.

## Common Use Cases

| Library             | Contents                                                       | Who Benefits      |
| ------------------- | -------------------------------------------------------------- | ----------------- |
| Company Handbook    | Values, process docs, HR policies                              | Everyone          |
| Engineering Library | API reference, architecture decisions, runbooks                | Engineering group |
| Competitor Research | Pricing pages, feature comparisons, news summaries             | Growth group      |
| Customer Context    | Key account notes, product requests, onboarding status         | CS team           |
| Brand Library       | Voice and tone guidelines, approved copy, messaging frameworks | Marketing group   |

## Creating a Library

1. In the Notara dashboard, navigate to **Libraries** in the sidebar.
2. Click **+ New Library**.
3. Give it a name and description.
4. Select the **group** it belongs to (or "Workspace" for a shared library).
5. Click **Create**.

You'll land on the library's document view, where you can start adding documents.

→ [Managing documents](/notara-docs/libraries/documents.md)

## The Manifest

Each library has an auto-generated manifest — a markdown index that the agent reads at the start of every session. It's regenerated automatically when documents are added, removed, or renamed.

You can view and manually regenerate the manifest from the library's header panel. The manifest includes:

* Library name and description
* Total document count
* A table of all documents: title, load policy, last updated, and a one-line summary

The agent reads the manifest first, then decides which `on_demand` documents to fetch based on the user's request. This is why the description and title of each document matter — they're what the agent uses to decide relevance.


---

# 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://notara.gitbook.io/notara-docs/libraries.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.