# Glean MCP for AI Agents AI Agent Connect > Glean MCP connects your enterprise knowledge base to your AI agent. It lets you search across Slack, Google Drive, Jira, and Confluence in one place. You can find people, manage documents, and get answers grounded in your actual company data without jumping between tabs. ## Overview - **Category:** productivity - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_lMfGFESPtGUfMbppNkUCSFtY6hpuymopiHHrJaK6/ai-agent-connect - **Tags:** enterprise-search, unified-search, people-discovery, ai-qa, knowledge-base, data-indexing ## Description Glean MCP connects your enterprise knowledge base to any AI agent. Instead of hunting through dozens of tabs to find a single policy or technical spec, you can just ask your agent to find it for you. It pulls data from your connected sources like Slack, Google Drive, Jira, and Confluence, giving you a single point of entry for everything your team knows. You can find specific documents, identify the right person to talk to for a project, or get answers grounded in your actual company data. When you use this with Vinkius, it acts as the bridge that lets your AI understand your organization's unique context. It handles the heavy lifting of searching through your corporate library so you can spend less time looking for information and more time actually using it to get work done. ## Tools ### chat Ask questions to the Glean AI to get answers grounded in your company's knowledge base. It's perfect for quick Q&A about internal policies. ### check_glean_status Verify that your connection to the Glean workspace is active. Use this to troubleshoot your setup quickly. ### delete_document Remove a specific document from the index. This keeps your knowledge base clean and up to date. ### get_collection Retrieve the details of a specific curated collection. Use this to see what's included in a set of docs. ### get_document Fetch the specific details of a single document. This helps you see metadata for a specific file. ### index_document Add a new document to the knowledge base. Use this to make sure your AI can see your latest files. ### list_collections View all the curated collections available in your workspace. It's a great way to see what's already organized. ### search_by_datasource Limit your search to a specific app like Slack or Confluence. This is useful when you know exactly where the info lives. ### search_people Find employees based on their name, role, or expertise. Use this to identify the right person to contact. ### search Perform a broad search across all connected data sources. This is the primary way to find anything in your company. ### autocomplete Get intelligent search suggestions as you start typing. This helps you find the right terms faster. ### bulk_index_documents Add a large batch of documents to the index at once. This is the best way to handle new folders or archives. ## Prompt Examples **Prompt:** ``` Find our deployment runbook and the on-call rotation. ``` **Response:** ``` I found the following relevant documents: | Document Name | Source | Last Updated | | :--- | :--- | :--- | | Production Deployment Runbook v3.2 | Confluence | 3 days ago | | On-Call Rotation Q2 2025 | Google Sheets | May 2 | Would you like me to summarize the rollback procedure from the runbook? ``` **Prompt:** ``` What is our refund policy for enterprise customers? ``` **Response:** ``` Based on the **Customer Policy Handbook** (Confluence, updated Feb 2025): * **30-Day Window:** Full refunds available within 30 days of contract signing. * **Prorated:** After 30 days, refunds are prorated based on remaining months. * **Approval:** Requires VP-level approval. Would you like to see the full policy document? ``` **Prompt:** ``` Who is the lead for the payments team? ``` **Response:** ``` The engineering lead for the Payments Team is **James Rodriguez**. * **Role:** Senior Engineering Manager * **Expertise:** Payment processing, PCI-DSS, Stripe integration * **Location:** San Francisco I can also search Slack for recent discussions involving James if you'd like. ``` ## Capabilities ### Search across all company apps Find anything in your connected data sources with a single query. ### Filter by specific platforms Limit your search results to just Slack, Drive, or other specific tools. ### Locate team members Find employees by their name, role, or expertise across the company. ### Manage document indexes Add new files or remove outdated content from your knowledge base. ### Browse curated collections Access specific sets of content like onboarding docs or company policies. ### Get grounded answers Ask questions and get answers based on your actual company data. ### Get search suggestions See relevant company terms and results as you start typing. ## Use Cases ### Finding a deployment runbook An engineer asks for the deployment runbook and gets the Confluence link and the Slack thread immediately. ### HR policy check An employee asks about the refund policy and gets a grounded answer from the handbook with a source link. ### Finding a team lead A new hire asks who leads the payments team and gets the name, location, and expertise of the lead. ### Bulk onboarding An IT admin uses bulk_index_documents to add 50 new training PDFs to a curated collection. ## Benefits - Stop switching tabs by using search to find info across all your company apps in one go. - Get accurate answers with chat because it uses your actual company data to ground its responses. - Find the right experts faster using search_people to see who has the specific skills you need. - Keep your knowledge base clean by using delete_document to remove outdated or sensitive files. - Speed up your workflow with autocomplete to see relevant company terms as you start typing a query. - Organize team knowledge by using bulk_index_documents to ingest entire folders of project specs at once. ## How It Works The bottom line is that it turns your fragmented company data into a single, searchable knowledge base for your AI. 1. Subscribe to the Glean MCP on Vinkius. 2. Enter your Glean API Token from the Admin Console. 3. Ask your AI agent to find documents, people, or answers. ## Frequently Asked Questions **Can Glean MCP search my Slack messages?** Yes, it connects to your Slack workspace to find relevant threads and messages based on your queries. **Does Glean MCP help with onboarding?** Yes, it can access curated collections specifically designed for onboarding and company policies. **Can I use Glean MCP to find people?** Yes, it lets you search for colleagues by their name, role, or specific area of expertise. **Is Glean MCP good for engineering teams?** It is excellent for engineers who need to quickly find technical specs, wikis, and tickets across Jira and Confluence. **How does Glean MCP handle my private data?** It uses your existing Glean workspace permissions, ensuring your AI agent only sees what you are authorized to see. **Can I see which app a result came from?** Yes, the results clearly identify whether the information came from Drive, Slack, Jira, or another connected source. **Can I search across multiple data sources like Confluence, Slack, and Google Drive simultaneously?** Yes! The `search` tool queries all connected data sources in a single request. If you want to narrow results to a specific platform, use `search_by_datasource` with the datasource name (e.g., 'Confluence', 'Slack', 'Google Drive'). Use `autocomplete` to get intelligent suggestions as you refine your query. **Can I ask the Glean AI assistant questions about my organization's knowledge?** Yes. The `chat` tool sends a natural language question to Glean's AI assistant, which generates an answer grounded in your organization's indexed knowledge — including documents, wikis, Slack conversations, and Jira tickets. This is ideal for getting synthesized answers rather than just search results. **Can I index custom documents into Glean programmatically?** Yes. Use `index_document` to add a single document with a JSON payload containing title, body, URL, and metadata. For large batches, use `bulk_index_documents` to index multiple documents in one request. Use `delete_document` to remove outdated content and `get_document` to verify indexing status.