Connect Google Slides to your AI chat agent
Client reports, pitch decks, recurring presentations — most of them start from the same template, rebuilt by hand every time.
SourcetoMCP connects Slides straight to Claude, ChatGPT, or any MCP-compatible agent, so it can read, build, and edit real slides instead.
Agent asks
What's on slide 4 of our pitch deck?
Agent asks
Duplicate our template slide for Acme Corp's Q3 numbers
Why Slides access matters for your agent
A deck your agent can only describe from a screenshot loses its structure. It needs real access to duplicate, edit, and build slides, not just talk about them.
- Built for templating. Design one slide by hand, then let your agent duplicate and fill it in per client, per record, per report.
- Full read and write, not just extraction. Read text and notes, then add, edit, or reorder slides in the same presentation.
- Precise, not approximate.Edit one specific shape or slide by ID, without a global find-and-replace touching content you didn’t mean to change.
How it works
Connect Google Slides. Authorize through Google's own sign-in, then pick the presentation with Google's own file picker. SourcetoMCP never sees your password.
Your agent asks in plain language. It calls the right MCP operation behind the scenes — reading slides, duplicating a template, or editing one shape — and gets the real deck back.
Every write shows exactly what changed. The slide or shape that was edited comes back with every write, so you can verify the result rather than take it on faith.
google_slides.get_presentation
returns slide_number, slide_id, text, speaker_notes
| Slide | Text |
|---|---|
| 3 | Q3 Highlights |
| 4 | Client Report Template |
| 5 | Next Steps |
Ask about your deck. Get a new slide built too.
‖ manual mode on
Built to read and build, not just describe slides
Read every slide
Text and speaker notes, slide by slide, in order — not a screenshot description.
Build decks from a template
Duplicate a slide and fill it in with find-and-replace, instead of recreating a deck by hand for every client or record.
Add and remove slides
Append a new slide with a real layout — title and body included — or delete and reorder existing ones.
Edit precisely
Update one specific shape's text, or a slide's speaker notes, without touching anything else on the page.
Add visuals
Place an image on any slide from a URL — logos, charts, photos.
See what you're working with
Pull a rendered thumbnail of any slide, or the presentation's real page size, before or after an edit.
No canned prompts needed, just ask
Google Slides doesn't have dedicated entries in the Prompt Library yet, but your agent doesn't need them. It already has direct read and write access to any connected presentation through the operations above, so you can ask in your own words from day one.
Google Slides MCP FAQ
Common questions from teams connecting Google Slides to an agent.
Read every slide's text and notes, duplicate/delete/reorder slides, append a new slide, edit a specific shape's text or a slide's notes, place an image, and pull a slide thumbnail — 12 operations in total, covering both reading and editing.
Claude, ChatGPT, Cursor, or any MCP-compatible client. Connect once from your dashboard's "Connect an agent" page and every tool you've paired can use it.
No. Connecting uses Google's own OAuth sign-in, and picking which presentation to connect uses Google's own file picker. SourcetoMCP never sees your password.
No. It only acts when you prompt it to, and every write returns exactly what changed, so you can see the result rather than take it on faith.
Design one slide by hand with {{placeholder}} tokens in the text. Your agent duplicates it per record, then replaces the tokens with real data on each copy — no rebuilding the layout each time.
Google Slides is available on every paid plan. See Pricing for how many connections each plan includes.