The server itself exposes 16 tools, and some of them can write to your Glean account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Glean.

How is this different from connecting Glean MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Glean MCP API changes?

We track the Glean MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Glean tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Glean?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Glean MCP server myself?

No. TrustGate connects to the Glean MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Glean directly.

Connect Glean MCP to your AI agents, securely.

One secured endpoint for every Glean call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Glean MCP now

What your agents can ask once Glean MCP is connected.

“Search for the latest version of our onboarding policy”
“What did we decide about the pricing change?”
“Read this document and summarise it”

“Who owns this service in the engineering team?”
“Find the code that handles this endpoint”
“Search my mailbox for the contract thread”

Select the tools your AI agents can call

16 tools from the Glean MCP schema.

Showing 16 tools

searchreadSearches Glean's content index across connected sources, returning only items the user can already access.
web_searchreadSearches the public web alongside company content.
code_searchreadSearches the indexed source code repositories.
meeting_lookupreadFind meeting recordings and notes.
gemini_web_searchreadSearch the web with Google Gemini.
knowledge_graph_queryreadQuery the Knowledge Graph for entities and their relationships.
knowledge_graph_schemareadRetrieve the Knowledge Graph schema, including entity types and predicates.
read_documentreadReads the full content of a document by ID or URL, including Library and Canvas content.
chatreadHolds a conversation with Glean's AI assistant, grounded in company content with citations.
memoryreadReads the memory Glean holds for the user across sessions.
memory_schemareadReturns the schema of the memory store, so an agent knows what it can retrieve.
data_analysisreadRoute analytical questions to the right BI tool or data warehouse.
employee_searchreadSearches the people directory by name, team or department.
user_activityreadRetrieves the authenticated user's recent activity across connected sources.
gmail_searchreadSearches the user's Gmail mailbox.
outlook_searchreadSearches the user's Outlook mailbox.

What Glean MCP exposes

Glean MCP groups its 16 tools into 4 categories, each with a suggested starting policy.

Category	Covers	Access	Suggested policy
Search	Enterprise search across connected sources, web search and code search	Read	Glean enforces per-document permissions; results stay within what the user can already see
Documents	Reading full document content by ID or URL, including Library and Canvas content	Read	Documents can hold confidential company content; restrict access
Chat & Memory	Conversational Q&A with the Glean assistant, plus memory and its schema	Read	Answers are grounded in indexed content; cite the sources
People & Activity	Employee directory search, user activity, calendar and mailbox search	Read	Mailbox and calendar reach personal data; add them only when the workflow needs them