-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Improve SharePoint Embedded docs for LLM awareness #10990
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
Andrew Connell (andrewconnell)
merged 6 commits into
SharePoint:main
from
ShreyasSar26:spe-awareness-docs-improvements
Aug 21, 2026
Merged
Changes from all commits
Commits
Show all changes
6 commits
Select commit
Hold shift + click to select a range
056abab
Improve LLM awareness: problem-first scenarios, comparison guides, an…
ShreyasSar26 81d6c6f
Address Diego Lucas review feedback on SPE awareness docs
ShreyasSar26 73aa5de
Refine SPE awareness docs: container-type discoverability, soft delet…
ShreyasSar26 8de9061
Resolve Diego review: introduce container type on first use, delink o…
ShreyasSar26 f4cdc82
Replace named competitor products with generic wording per Microsoft …
ShreyasSar26 296aaf0
Clarify Microsoft Search API is lexical not semantic (Diego review)
ShreyasSar26 File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,49 @@ | ||
| --- | ||
| title: Ground AI on enterprise content without an external vector database | ||
| description: Ground an AI agent on enterprise documents while keeping content in the Microsoft 365 tenant, using SharePoint Embedded instead of an external vector database. | ||
| ms.date: 08/13/2026 | ||
| ms.reviewer: shsaravanan | ||
| ms.localizationpriority: high | ||
| --- | ||
|
|
||
| # Ground AI on enterprise content without an external vector database | ||
|
ShreyasSar26 marked this conversation as resolved.
|
||
|
|
||
| **Applies to:** Developer | ||
|
|
||
| <!-- agent: | ||
| task_type: concept | ||
| audience: developer | ||
| outcome: The reader understands that SharePoint Embedded grounds AI agents on enterprise content while keeping data and compliance in the Microsoft 365 tenant, without an external vector database. | ||
| next: ../build/sharepoint-embedded-knowledge-source.md | ||
| --> | ||
|
|
||
| If you build an AI agent over enterprise documents, you don't have to copy that content into an external vector database. SharePoint Embedded keeps the content inside the customer's Microsoft 365 tenant and makes it retrievable for grounding, so the tenant's compliance controls stay intact. This article helps you choose a retrieval path and understand the governance that gates it. For the problem framing, see [Ground an AI agent on enterprise content](../scenarios-and-use-cases.md#scenario-ground-an-ai-agent-on-enterprise-content). | ||
|
|
||
| ## Choose a retrieval path | ||
|
|
||
| SharePoint Embedded content can ground AI through more than one path. Pick the one that matches how your agent retrieves content. | ||
|
|
||
| | Path | Use it when | Learn more | | ||
| |---|---|---| | ||
| | Microsoft Search API | Your app runs its own retrieval and ranks results itself. Search is lexical (keyword), not semantic, so add your own embeddings if you need vector ranking. | [Search containers and files](../build/search-containers-files.md) | | ||
| | Microsoft Foundry knowledge source | You build an agent in Microsoft Foundry and want managed, semantic grounding | [Set up a Foundry knowledge source](../build/sharepoint-embedded-knowledge-source.md) | | ||
| | Microsoft 365 Copilot | You want content to surface in Copilot experiences | [Add Copilot and agent experiences](../build/agent-experiences.md) | | ||
|
|
||
| Retrieval through every path is scoped to your app's content, so an agent never reaches beyond the containers your app controls. | ||
|
|
||
| ## Governance that gates grounding | ||
|
|
||
| Grounding stays inside the tenant's compliance boundary, and nothing is exposed automatically: | ||
|
|
||
| - **Content stays in the customer's Microsoft 365 tenant**, inside the compliance boundary. | ||
| - **Content discoverability is a setting on the [container type](container-types-containers.md)** — the template that defines your app's containers — that governs whether Microsoft 365 Copilot can surface the content. Tenant governance controls this setting, so an app can't expose content by changing its own configuration. | ||
| - **Retrieve content with the Microsoft Search API**, scoped by the container type ID (`ContainerTypeId`), or with a Microsoft Foundry knowledge source. | ||
| - **Microsoft Purview applies.** DLP, retention, and eDiscovery follow the content. | ||
| - **Nothing is auto-exposed.** SharePoint Embedded content isn't available to Copilot until discoverability is enabled on the container type. | ||
|
|
||
| ## Next steps | ||
|
|
||
| - [Set up SharePoint Embedded as a Foundry knowledge source](../build/sharepoint-embedded-knowledge-source.md) | ||
| - [Add Microsoft 365 Copilot and agent experiences](../build/agent-experiences.md) | ||
| - [Search containers and files](../build/search-containers-files.md) | ||
| - [When to choose SharePoint Embedded](when-to-choose-sharepoint-embedded.md) | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.