> ## Documentation Index
> Fetch the complete documentation index at: https://docs.intelligence-management-platform.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Pinned knowledge and presets

> Pin knowledge contexts or single items to a conversation, and save selections you reuse as named presets.

export const RightsCallout = ({right, flags}) => <Info>
    Who sees this: requires {right === "none" ? "no special rights — available to every signed-in user" : `the ${right} right (or super admin)`}.
    {flags && flags !== "none" ? ` Only visible when the server enables ${flags}.` : ""}
  </Info>;

<RightsCallout right="none" flags="none" />

Pinning knowledge tells the agent where to look. Pin a whole context (for example an HR handbook) or individual items (specific documents) to a conversation, and the agent's searches focus on what you pinned. The terms *context* and *item* are defined in [Core concepts](/get-started/concepts).

## Pin knowledge to a conversation

<Frame caption="Pinning knowledge: ＋ → Add knowledge opens the Add Context & Items picker — select an item and it appears as a chip above the composer.">
  <video autoPlay muted loop playsInline src="https://mintcdn.com/exulu/HUjjVgs8aB3PSMha/videos/pin-knowledge.mp4?fit=max&auto=format&n=HUjjVgs8aB3PSMha&q=85&s=b6c6a362e9a959acea0cf5dcce2ec7b1" data-path="videos/pin-knowledge.mp4" />
</Frame>

<Steps>
  <Step title="Open the knowledge picker">
    Click **＋** in the composer and choose **Add knowledge** ("Pin contexts or items for this chat").
  </Step>

  <Step title="Pick contexts or items">
    Browse or search the contexts you have access to. Select a whole context to pin everything in it, or open a context and pick individual items.
  </Step>

  <Step title="Check the chips">
    Pinned contexts and items appear as chips under the message box. Click a chip's **×** to unpin it. The pins stay with the conversation — they apply to every message you send in it.
  </Step>
</Steps>

<Note>
  Some agents run with a managed context: "This agent only searches contexts and items you add to this session." With such an agent, pinning isn't optional focus — it defines everything the agent can search, so pin before you ask.
</Note>

## Save a selection as a context preset

If you pin the same combination repeatedly, save it once:

<Steps>
  <Step title="Save the current selection">
    With your contexts and items pinned, click **Save context preset** at the end of the chip row. The **Save as context preset** dialog opens.
  </Step>

  <Step title="Name and describe it">
    Give the preset a **Name** (required — for example "Customer Support Knowledge Base") and optionally a **Description** ("Describe when to use this preset...") and **Tags**.
  </Step>

  <Step title="Save">
    Click **Save preset**. You'll find it in the presets tab of the **Add knowledge** dialog from now on.
  </Step>
</Steps>

## Apply a preset

Open **＋ → Add knowledge**, switch to the presets tab, and apply a preset. Applying replaces the conversation's current pins with the preset's selection, and a preset chip appears at the start of the chip row.

If you then pin or unpin something, the chip shows "modified":

* **Update preset** saves your changed selection back into the preset (only shown if the preset is yours or shared with you for editing).
* The **×** on the preset chip deselects the preset without touching your current pins.

## Share a preset

Presets are private by default ("Using default permissions (private to you)"). To share one, open **Sharing & permissions** in the preset's save or edit dialog and choose who can view and use it — specific users, roles, or teams.

<Tip>
  Shared presets are how a team standardizes on the "right" knowledge for a job: one person curates the selection, everyone else applies it with two clicks.
</Tip>