Sign in Book a demo
Back to Tools
Search OAuth BYOC
Build Your Agent View Glean Documentation
glean logo

Glean MCP Tools for AI Agents

Let an agent search a user's Glean enterprise index and return ranked results from that user's authorized account.

Available Tools

2 tools available in this integration

Glean.Search

Search the user's Glean enterprise index and return ranked results.

Glean.SearchDocuments

Search the user's Glean enterprise index and return ranked results.

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "find the current security policy for handling customer data and cite the source documents." Arcade picks the right Glean searches without guessing document IDs or search syntax.

Up to 100x fewer tokens

Glean search returns ranked results in one response, so the agent can compare relevant documents and citations without loading the entire enterprise index into its context.

Complete the workflow end to end

Search the enterprise index, compare ranked results, identify the most relevant documents, summarize the findings, and return source references through one toolkit designed around the complete knowledge-search task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing Glean

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Glean, you can implement some of these policies:

Access when tools are listed for a user
  • Decide which tools an agent is allowed to see at all. A tool that is never listed cannot be called, whoever is asking.
Pre-execution before each tool call
  • Inspect the inputs a call is about to run with, and block it or send it for human approval when they fall outside what the agent should do unsupervised.
Post-execution after the tool returns
  • Redact document bodies, snippets, and personal contact details from Glean.Search results before they reach the model, so an agent cannot expose private enterprise content in its response.

How to use Glean with your agent

Pick your MCP client below and follow the walkthrough to connect an agent to Glean with per-user OAuth. Your agent can then search the enterprise index on behalf of each authorized user.

Your questions, answered

A Glean MCP server exposes Glean search as a structured capability that an AI agent can call through the Model Context Protocol. After a user authorizes their Glean account, the agent can search that user's enterprise index and receive ranked results.

Arcade provides Glean search through its MCP gateway and adds per-user OAuth, approval policies, response redaction, and credential brokering. This gives agents and clients a consistent way to reach each user's authorized Glean index without building the connection separately.

Arcade agents can search the user's Glean enterprise index and receive ranked results. The available actions are read-only: they do not create, update, delete, or share Glean content.

Each user authorizes their own Glean account through custom OAuth2. The granted authorization controls which enterprise-indexed content that user can search; grant only the access needed for the workflow.

Yes. Each user connects their own Glean account through OAuth, and searches run under that user's authorization. An agent can access only the Glean content available to that user.

The available Glean actions are read-only searches, so they do not perform consequential changes such as editing or deleting content. Contextual Access policies can still require approval before a search call, redact fields from its response, or hide the capability from an agent entirely.

Arcade proxies Glean API calls and does not persist your content as part of normal tool execution. Credentials are brokered and stored encrypted at rest, and the model never receives them.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade refreshes access while the user's Glean authorization remains valid. If that authorization is revoked, rotated, or can no longer be refreshed, searches stop until the user reconnects Glean.

Glean's API rate limits and account-level quotas apply to searches made through Arcade. The available Glean search actions do not add a separate quota commitment.

Ready to actually ship your agent?

Connect your agent to Glean with managed OAuth, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the Glean Quickstart