Google Slides MCP Server: Secure AI Agent Access to Google Slides
Connect Claude, ChatGPT, and Cursor to Google Slides through PortEden's remote MCP server. Read deck content and speaker notes over OAuth, with a data firewall on every call.
Free tier · No credit card required
What is the Google Slides MCP server?
The Google Slides MCP server lets AI agents read presentations through the Model Context Protocol. Connect a client like Claude, ChatGPT, or Cursor to one remote endpoint at https://mcp.porteden.com/google-slides and the agent can pull a deck's outline, slide text, and speaker notes, with no copy and paste.
PortEden's server reads presentation metadata (the deck title and a navigable list of slides) and presentation content, either as plain text with speaker notes or as the raw structured Google Slides JSON. What makes it different from a generic Google Slides MCP is the data firewall in front of those reads: PortEden inspects every tool-call request, applies your access policy, and redacts sensitive content in the response before the agent ever sees it.
The result is secure Google Slides access for AI agents. You decide which presentations a client can read, redact sensitive text on slides and in speaker notes, and review every call in an audit log. OAuth handles authentication, so there is no static key to leak. Presentation files live in Drive, so finding, sharing, and managing decks is handled by the Drive MCP server.
Google Slides MCP tools
Every tool can be allowed or denied per client. Access levels are enforced at the firewall on each call.
These are the 2 documented read tools, both gated by the read_slide_content operation flag. PortEden's live Google Slides server exposes 7 tools; finding, sharing, and managing the underlying presentation files is handled by the Drive server. Newer tools will appear here as they are documented.
| Tool | Access | What it does |
|---|---|---|
google_slides_get_metadata | read | Get a presentation's title and a list of its slides, each with its index and slide title. Use it to plan a content fetch or to show a navigable slide list. |
google_slides_read | read | Read a presentation's content as plain text (slide bodies plus speaker notes) or as structured Google Slides JSON (slide object IDs, page elements, layouts, and masters). |
Connect Google Slides to Claude, ChatGPT, Cursor, and more
Point any MCP-compatible client at the remote URL and sign in to PortEden once with OAuth.
# Claude Web or Desktop, then Settings, Connectors, Add custom connectorhttps://mcp.porteden.com/google-slides # Authenticate to PortEden once with OAuth. Claude can now read your# presentations under the access policy you set.Available on Claude Pro, Team, and Enterprise. The same flow works for Claude Cowork.
The security-first Google Slides MCP
PortEden is the data firewall for AI.
Granular access control
Decide exactly which presentations an AI client can read. Both Slides tools are gated by the read_slide_content operation flag, so you can scope a client to specific decks or deny slide reading entirely while still allowing other access.
- Read-only by design: the Slides server only reads, so an agent can never edit or delete a deck through it.
- Per-presentation scope: limit a client to a named set of decks and block the rest.
- Deny slide reading: turn off read_slide_content for an agent that should never see presentation contents.
Data redaction
PortEden runs a redaction pass on every tool response before it leaves the firewall. Sensitive text on slides and in speaker notes is replaced with stable placeholders, so the agent can summarize a deck without the raw content entering the model's context.
- Strip PII, financial figures, and unreleased product names from slide bodies before the agent sees them.
- Redact speaker notes, which often hold the most candid context, while keeping the on-slide outline readable.
- Redaction applies to both the text and the structured response formats.
Audit trail
Every Slides tool call is recorded: which client, which user, which tool, the arguments passed, the policy decision, and the redacted response. Export it or stream it to your SIEM for review.
- See exactly which decks an agent read and when, with timestamps.
- Reconstruct any tool call: the request, the rule that fired, and the response returned.
- Stream to Splunk, Datadog, or S3 for retention and review.
RBAC and policy groups
Bind each MCP connection to a user or role with a scoped, revocable token. Group policies by team so a new hire inherits the right presentation access on day one, and revocation is instant and server-side.
- Issue per-user tokens scoped to the presentations that role needs.
- Apply one policy group across a team instead of editing rules client by client.
- Revoke a token server-side the moment a contract ends, with no Google account round-trip.
Set up the secure Google Slides MCP in minutes
Add the connector
In your AI client, add a custom connector or HTTP MCP server pointing at https://mcp.porteden.com/google-slides.
Authenticate with OAuth
Sign in to PortEden once and connect your Google account. The client never holds your Google refresh token.
Set your Slides policy
Choose which presentations a client can read, and turn on redaction for sensitive slide text and speaker notes.
Verify in the audit log
Run a prompt, then watch the tool calls land in your PortEden audit log with the rule that fired on each one.
Google Slides MCP FAQ
What is the Google Slides MCP server?
How do I connect Claude to Google Slides using MCP?
What can the Google Slides MCP server do?
Can the Google Slides MCP server read speaker notes?
Is the Google Slides MCP server secure?
Can I stop an agent from reading certain presentations?
What Google Slides tools does the MCP server expose?
Which AI clients work with the Google Slides MCP server?
Connect Google Slides to AI, without leaking the underlying data.
Five-minute setup over OAuth. The free tier covers 1,000 tool calls per month.