Integration

SlideSpeak integration

SlideSpeak generates PowerPoint presentations with AI. v1 (template-based): generate a deck from a topic, documents or an outline, generate an outline, build a deck slide by slide, insert, regenerate or remove a slide and roll an edit back, fill named shapes of your own .pptx. v2 (next-generation models): generate a deck with a design kit and reusable sections, import a .pptx, translate it, turn it into a narrated video. Generations spend API credits (v1: 1 per slide; v2: 5 per slide or video slide, 1 per translated slide). Authentication is an API key (app, Settings, Developer) sent in the X-API-Key header.

Connect SlideSpeak How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in SlideSpeak.Built from the 27 SlideSpeak tools below, on the connection you add once.

What you can do with SlideSpeak

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Download, 2 operations

get_download_urlGet download urlGETReturns url, a short-lived link to the file of a finished v1 presentation, given its request_id. The link expires soon; call again for a fresh one.
get_download_url_v2Get download url v2GETReturns url, a short-lived link to the file (PowerPoint, or MP4 for a video) of a finished v2 task, given its request_id. The link expires soon; call again for a fresh one.

Generate, 2 operations

generate_presentationGenerate presentationPOSTGenerates a v1 deck with AI from a topic, uploaded documents or an outline, on a template. Spends 1 credit per slide; returns request_id for get_download_url. Waits for the job and returns the finished result; gives up after 10 minutes.
generate_presentation_v2Generate presentation v2POSTGenerates a deck from a prompt with the v2 models (documents, design kit optional). Spends 5 credits per slide; request_id goes to get_download_url_v2. Waits for the job and returns the finished result; gives up after 10 minutes.

Me, 2 operations

get_accountGet accountGETReturns the user name behind the API key and its remaining API credits (api_credits and additional_api_credits). get_account_v2 is the same read on the v2 API.
get_account_v2Get account v2GETReturns the user name behind the API key and its remaining API credits (api_credits and additional_api_credits), through the v2 API. get_account is the same read on v1.

Subscribe, 2 operations

create_webhookCreate webhookPOSTSubscribes a URL to POST notifications when a presentation generation task completes; returns webhook_id. No signing secret is documented, so deliveries cannot be verified.
create_webhook_v2Create webhook v2POSTSubscribes a URL, through the v2 API, to POST notifications when a presentation generation task completes; returns webhook_id. No signing secret is documented, so deliveries cannot be verified.

Task status, 2 operations

get_task_statusGet task statusGETReturns the state of a v1 task: task_status (SENT or STARTED while running, SUCCESS, FAILURE), task_result and task_error. For a job a generation tool stopped waiting for.
get_task_status_v2Get task status v2GETReturns the state of a v2 task: task_status (SENT or STARTED while running, SUCCESS, FAILURE), task_result and task_error. For a job a generation tool stopped waiting for.

Unsubscribe, 2 operations

delete_webhookDelete webhookDELETEDeletes a webhook subscription by webhook_id: notifications to its endpoint stop.
delete_webhook_v2Delete webhook v2DELETEDeletes a webhook subscription by webhook_id, through the v2 API: notifications to its endpoint stop.

Upload, 2 operations

upload_documentUpload documentPOSTUploads a document (.pptx, .ppt, .docx, .doc, .xlsx, .pdf, .txt, .md) for v1 and waits for processing; task_result is its id for document_uuids. Waits for the job and returns the finished result; gives up after 10 minutes.
upload_document_v2Upload document v2POSTUploads a document (.pptx, .ppt, .docx, .doc, .xlsx, .pdf, .txt, .md) for v2 and waits for processing; task_result is its id for document_ids. Waits for the job and returns the finished result; gives up after 10 minutes.

Branded, 1 operation

list_branded_templatesList branded templatesGETLists the account's processed branded templates: id (pass it as template to generate_presentation or generate_presentation_slide_by_slide) and file name.

Design kit, 1 operation

list_design_kitsList design kitsGETLists the design kits available to the account: id (pass it as design_kit_id to generate_presentation_v2), name and preview image URLs.

Edit, 1 operation

replace_shape_contentReplace shape contentPOSTFills a PowerPoint file you upload: replaces the content of named shapes (config.replacements, each shape_name and content) and returns url, a link to the edited .pptx.

Import, 1 operation

import_presentationImport presentationPOSTImports a .pptx file (at most 100 MB) into the account and returns its presentation_id, for translate_presentation or generate_presentation_video. Free.

Outline, 1 operation

generate_outlineGenerate outlinePOSTGenerates a slide outline with AI, or edits one (outline plus prompt); 1 credit per 5 slides. Pass task_result.slides as outline to generate_presentation. Waits for the job and returns the finished result; gives up after 10 minutes.

Reusable section, 1 operation

list_reusable_sectionsList reusable sectionsGETLists the account's reusable sections (saved slides inserted as-is): id (pass it as reusable_section_id to generate_presentation_v2), name, placement and slide_count.

Rollback, 1 operation

rollback_editRollback editPOSTRestores a presentation to its state before an earlier edit, given that edit's previous_version_id. The whole deck is restored, web editor changes included. Free. Waits for the job and returns the finished result; gives up after 10 minutes.

Slide, 1 operation

edit_slideEdit slidePOSTChanges a v1 presentation: inserts, regenerates or removes the slide at position. INSERT and REGENERATE spend 1 credit; previous_version_id undoes it with rollback_edit. Waits for the job and returns the finished result; gives up after 10 minutes.

Slide by slide, 1 operation

generate_presentation_slide_by_slideGenerate presentation slide by slidePOSTGenerates a v1 deck from slides you define one by one (layout, item_amount, content). Spends 1 credit per slide; request_id goes to get_download_url. Waits for the job and returns the finished result; gives up after 10 minutes.

Template, 1 operation

list_templatesList templatesGETLists SlideSpeak's default templates: each name (pass it as template to generate_presentation or generate_presentation_slide_by_slide) with cover and content preview image URLs.

Translate, 1 operation

translate_presentationTranslate presentationPOSTTranslates a presentation into another language as a new PowerPoint file; the original is unchanged. Spends 1 credit per slide; returns request_id for get_download_url_v2. Waits for the job and returns the finished result; gives up after 10 minutes.

Video, 1 operation

generate_presentation_videoGenerate presentation videoPOSTTurns a presentation into a narrated MP4 video (takes one to several minutes). Spends 5 credits per slide; returns request_id for get_download_url_v2. Waits for the job and returns the finished result; gives up after 10 minutes.

Voice, 1 operation

list_voicesList voicesGETLists the narration voices for a language (voice_id, name, gender, Azure locale), to pick one for generate_presentation_video. Free.

Use SlideSpeak from Claude Code, Cursor or Codex

Connect SlideSpeak in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its SlideSpeak operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect SlideSpeak

Create your workspace
Sign up free. Nothing to install.
Connect SlideSpeak once
Add your SlideSpeak connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.