​

slides

orc slides selects a slides backend with --target CONNECTION_ID. Google Workspace is the currently supported backend. The connection ID is a flag.

get returns the presentations.get response; create accepts --title. api get and api update call read and write methods from the pinned Discovery document. batch-requests create validates exactly one Request key per object. schemas list/get inspect methods, Request types, and schemas without a connection.

Reconnect existing Workspace connections to grant Slides scopes. Writes require manage access. --readonly blocks writes; --wrap-untrusted wraps external text. After a timeout, read the presentation before retrying a write.

exports get PRESENTATION_ID --format pdf|pptx --output FILE saves PDF or PowerPoint (Drive export, maximum 10 MB). The drive.file grant covers files created by the app or explicitly opened with it through Picker or a similar app file picker. User ownership alone is insufficient; sharing with the account or reconnecting does not itself grant app access to every existing file. Inaccessible files return CONFLICT with share/reconnect guidance. Orchestor does not request the restricted drive or drive.readonly scopes.

thumbnails get PRESENTATION_ID PAGE_ID [--size SMALL|MEDIUM|LARGE] returns content_url, width, and height. With --output FILE, the CLI downloads the PNG. The URL normally lasts 30 minutes; anyone with it can access the image. Both commands work with --readonly.

create --markdown FILE creates a deck from Slidev Markdown; update PRESENTATION_ID --markdown FILE replaces every slide. FILE is a UTF-8 file path, or - for raw stdin. API clients send the content in markdown. Headmatter title names the presentation. Creating with only --title remains supported. Titles, body text, notes, and hide are converted. Unsupported features use safe plain text fallbacks and appear in report with a zero-based slide_index.

get PRESENTATION_ID --format markdown [--output deck.md] converts an existing presentation to Slidev Markdown. It retains titles, body text, lists, tables, image URLs, speaker notes and skipped slides, and reads hand-added text in position order. The API returns { markdown, report }. Layout aliases normalize to cover or center where Google cannot distinguish them. Unsupported shapes, videos, charts and other elements appear in report, as do expiring contentUrl references. Images prefer sourceUrl.

Get a presentation

Create a presentation

Get a presentation

Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Required.

Create a presentation

Path to a UTF-8 file, or - for stdin. Slidev Markdown content. CLI: path to a UTF-8 file, or - for stdin.; max 1000000 chars

Type: string. Optional.

(required) Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Optional.

Body field: title; max 500 chars

Type: string. Optional.

Call a read-only Slides API method

(required) Read-only Discovery method ID, for example presentations.pages.get.; enum: presentations.get|presentations.pages.get|presentations.pages.getThumbnail

Type: string. Optional.

Path and query parameters by Discovery name, for example {"presentationId":"…","pageObjectId":"…"}. Run orc slides schemas get <method> for parameters.; (JSON object, e.g. '{"custom_id":"x"}')

Type: string. Optional.

(required) Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Optional.

Call a mutating Slides API method

Request body in the Discovery request schema.; (JSON object, e.g. '{"custom_id":"x"}')

Type: string. Optional.

(required) Mutating Discovery method ID, for example presentations.batchUpdate.; enum: presentations.batchUpdate|presentations.create

Type: string. Optional.

Path and query parameters by Discovery name, for example {"presentationId":"…","pageObjectId":"…"}. Run orc slides schemas get <method> for parameters.; (JSON object, e.g. '{"custom_id":"x"}')

Type: string. Optional.

(required) Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Optional.

Apply Slides batchUpdate requests

(required) batchUpdate requests, each with exactly one Request key, for example {"createSlide":{…}}. Run orc slides schemas list for types.; JSON array of objects (use --stdin for large resources)

Type: string. Optional.

(required) Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Optional.

Slides WriteControl, for example {"requiredRevisionId":"…"}.; (JSON object, e.g. '{"custom_id":"x"}')

Type: string. Optional.

List Slides API methods and request types

Get a Slides API schema

Export a presentation

Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Required.

Get a slide thumbnail

Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Required.

Thumbnail width: SMALL 200px, MEDIUM 800px, LARGE 1600px. Google chooses the default when omitted.; enum: SMALL|MEDIUM|LARGE

Type: string. Optional.

Replace all slides from Slidev Markdown

(required) Path to a UTF-8 file, or - for stdin. Slidev Markdown content. CLI: path to a UTF-8 file, or - for stdin.; max 1000000 chars

Type: string. Optional.

(required) Connection ID of the slides backend (currently a google-workspace connection); max 200 chars

Type: string. Optional.

Inspect a method schema

Read a page

Apply batch requests

Export PDF

Export PowerPoint

Save a slide thumbnail

Create from Markdown

Replace all slides from stdin

Read as Slidev Markdown

For Markdown, --dry-run sends dry_run: true to the API and performs no Google writes. Create does not call Google; update reads the existing slide IDs and revision. The response includes presentation_id (null for create), slide_count, requests (Phase 1), notes (slide_object_id and text), and report. Update also returns write_control.requiredRevisionId. Notes are deferred text, not requests, because their object IDs exist only after slide creation. Real execution writes slides in Phase 1, then only when notes exist reads their IDs and a fresh revision and writes all notes in Phase 2. A notes failure reports that slides were already written. Re-run update to replace the slides and retry notes.

Lists become native bullets or numbering, including nesting. Each top-level list uses one preset throughout; mixed nested list types use the top-level style and add a lists report entry. Tables become editable tables with a bold header row. Bold, italic, inline code (Courier New), and links become text styles, with UTF-16 ranges. Markdown images and frontmatter image: accept absolute HTTPS URLs; Google fetches them, and Orchestor never downloads or uploads them. background: sets a stretched page image. Relative, data:, and HTTP sources appear in report; other content still converts. Mermaid, KaTeX, code blocks, v-click, components, and unknown options keep report entries. Ordinary code blocks remain editable monospaced text.

Image placement uses a 16:9 page (9,144,000 × 5,143,500 EMU). Mixed text, tables, and Markdown images stack within each body or column; the body placeholder is replaced with explicitly sized native elements when needed. HTTPS images must be accessible to Google. createImage preserves aspect ratio; background images stretch.

Requires a Workspace and connector:backend-execute. Schema commands need no connection.

The following global options can be used with orc slides:

For details and examples, see global options.