Auto-Generate OpenAI Embeddings for SharePoint Documents to Power Search

The agent scans a SharePoint document library for new or recently updated files, extracts their text content, generates embeddings via OpenAI, and stores them in a vector store to enable semantic search across the knowledge base.

How the workflow runs

The scenario uses specific integration tools at each step, while Metorial keeps access scoped and visible.

  1. 1

    Discover new and updated documents

    Search the SharePoint document library for files added or modified since the last indexing run, checking version history to avoid reprocessing unchanged files.

    • microsoft-sharepoint:search_drive
    • microsoft-sharepoint:get_file_versions
  2. 2

    Extract and embed document content

    Download each qualifying document, extract its text, and generate vector embeddings using OpenAI's embedding model.

    • microsoft-sharepoint:manage_file
    • openai:create_embeddings
  3. 3

    Update vector store and validate indexing

    Store the new embeddings in the OpenAI vector store and run a test semantic search to confirm the documents are retrievable.

    • openai:create_vector_store
    • openai:search_vector_store

Integrations used in this scenario

microsoft-sharepoint

Search Drive

Find recently added or modified files in the document library.

View details

microsoft-sharepoint

Manage File

Download the content of each identified document.

View details

microsoft-sharepoint

Get File Versions

Check version history to process only genuinely updated files.

View details

openai

Create Embeddings

Generate vector embeddings for the extracted document text.

View details

openai

Create Vector Store

Initialize or update the vector store for the knowledge base.

View details

openai

Search Vector Store

Validate that newly indexed documents are retrievable via semantic search.

View details

Connected systems

Integration

Sharepoint

Manage SharePoint sites, document libraries, lists, and files. Create, read, update, and delete lists and list items with custom columns. Upload, download, move, copy, and version files in document libraries. Search across sites, files, folders, lists, and list items using Microsoft Search. Manage permissions at site, list, and item levels with granular access control. Define and manage content types and site columns. Subscribe to webhooks for list and library change notifications. Retrieve site properties and search for sites across Microsoft 365.

View Sharepoint

Integration

Openai

Generate text, images, audio, and video using large language models and multimodal AI. Create chat completions, generate and edit images from text prompts, convert text to speech, transcribe and translate audio, generate video, and create text embeddings for search and retrieval. Fine-tune models on custom training data, run evaluations to measure model performance, and moderate content against policy categories. Manage vector stores for semantic file search, upload and organize files, and submit batch processing jobs for asynchronous bulk requests. Conduct real-time speech-to-speech conversations via WebRTC or SIP. Administer organizations, projects, users, API keys, and audit logs programmatically. Receive webhook notifications for background responses, batch jobs, fine-tuning jobs, eval runs, and incoming realtime calls.

View Openai

Expected outcomes

Outcome 1

SharePoint knowledge bases become semantically searchable without custom infrastructure

Metorial keeps the workflow connected, governed, and traceable across the systems involved.

Explore scenarios

Outcome 2

New documents are indexed automatically as they are uploaded

Metorial keeps the workflow connected, governed, and traceable across the systems involved.

Explore scenarios

Outcome 3

Teams can retrieve relevant documents using natural language queries

Metorial keeps the workflow connected, governed, and traceable across the systems involved.

Explore scenarios

How Metorial powers this scenario

Metorial is the governed connection layer between your AI agents and the tools your company runs on. It turns workflows like auto-generate openai embeddings for sharepoint documents to power search into something you can deploy quickly, safely, and at scale.

Fast

Ready for your entire team

Connect 1000+ verified integrations through one Magic MCP URL instead of building and maintaining bespoke connectors for each system in this workflow.

Browse integrations

Secure

Guardrails on every action

Protoguard inspects every message and tool call for prompt injection and policy violations before an agent touches your systems.

See how Protoguard works

Enterprise

SSO, policies, and audit trails

Agents act on real identity under company SSO, with per-user and per-group access policies and a complete, searchable record of everything that happens.

Explore enterprise

Team ready

Reusable across your org

Package this workflow as a skill, attach the tools it needs, and let teammates run it through Portals — governed by admins, owned by the people who do the work.

See Skills & Portals

Products behind this workflow

The Metorial products that connect, govern, and observe this scenario.

Connectivity

Integrations

Start from 1000+ verified integrations or bring your own, and give every one a governed path to your agents under existing SSO and access policies.

Explore Integrations

Connectivity

Magic MCP

A single URL your AI client connects to. Sign in with the login you already use and your agent reaches every integration and tool you allow — no per-app setup.

Explore Magic MCP

Identity

Access Control

Sign in with company SSO, set policies per user and group, and let agents act on real identity across every connected system in this workflow.

Explore Access Control

Governance

Protoguard

Metorial’s security layer reviews every message and tool request before an agent acts — catching prompt injection and blocking anything outside your policies.

Explore Protoguard

Observability

Tracing

A complete, searchable record of everything your agents, team, and machines do across these integrations, so you can trust the workflow in production.

Explore Tracing

Governance

Portals

Let teammates connect agents to the integrations and skills your company already uses, with admins deciding who gets access to what.

Explore Portals

Built for your whole team

However you adopt AI, Metorial has a path for connecting it safely.

Solution

For Agents

Give the agents behind this scenario governed access to every tool and integration they need, with one connection layer instead of bespoke glue code.

Agents solution

Solution

For Enterprise

SSO, granular access control, security review, and full audit trails so this workflow meets enterprise governance and compliance requirements.

Enterprise solution

Solution

For your Workforce

Let the people who do this work connect their own AI agents to approved integrations and reusable skills — safely, without waiting on engineering.

Workforce solution

Build this workflow with your own tools

Metorial gives teams one governed layer for connecting integrations to real production work.