Adobe MCP server for AI agents
Let Claude, ChatGPT or your own agent read and act on Adobe: 15 typed functions through one MCP server. No Adobe API key needed.
- Site
- adobe.com
- Functions
- 15
- Sign-in
- For account actions
- Works with
- MCP, HTTP, Node, Python
Connect
An MCP server for Adobe alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.
https://api.bowmark.ai/mcp/site/adobeOne MCP server for Adobe and every other site Bowmark covers. Your agent looks up what it needs and calls it.
https://api.bowmark.ai/mcpSet it up in your clientBoth need a Bowmark API key. The quickstart gets you one and a first result in a few minutes. Functions that need a login use the person's own account: how sign-in works.
Functions
Open a function to see its arguments, what it returns and how to call it.
compressPdfRuns Adobe Acrobat's free online "Compress PDF" tool (acrobat.adobe.com/.../compress-pdf.html) on a caller-supplied PDF and returns a smaller PDF at…Arguments
Name Type Required pdfUrlstringA public https URL of the PDF to compress (up to 5 MB, the most Adobe's guest upload takes).Yes level"low" | "medium" | "high""low" keeps the most quality, "high" the smallest file; defaults to "medium".No Returns
AcrobatPdfResultinterface AcrobatPdfResult { name: string; mimetype: string; /** A signed Adobe link to the result; short-lived, so fetch or save it promptly. */ downloadUrl: string; /** Size of the result in bytes. */ bytes: number; /** Combined size of the input PDFs in bytes. */ inputBytes: number; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/compressPdfCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.compressPdf({ pdfUrl: "<pdfUrl>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.compressPdf({"pdfUrl": "<pdfUrl>"}))curl -X POST https://api.bowmark.ai/v1/sites/adobe/compressPdf \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"pdfUrl":"<pdfUrl>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.compressPdf({ pdfUrl: "<pdfUrl>" });convertPdfToWordRuns Adobe Acrobat's free online PDF-to-Word converter on a caller-supplied PDF and returns an editable .docx, the way a visitor using the tool in a…Arguments
Name Type Required pdfUrlstringA public https URL of the PDF to convert (up to 5 MB).Yes Returns
AcrobatPdfResultinterface AcrobatPdfResult { name: string; mimetype: string; /** A signed Adobe link to the result; short-lived, so fetch or save it promptly. */ downloadUrl: string; /** Size of the result in bytes. */ bytes: number; /** Combined size of the input PDFs in bytes. */ inputBytes: number; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/convertPdfToWordCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.convertPdfToWord({ pdfUrl: "<pdfUrl>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.convertPdfToWord({"pdfUrl": "<pdfUrl>"}))curl -X POST https://api.bowmark.ai/v1/sites/adobe/convertPdfToWord \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"pdfUrl":"<pdfUrl>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.convertPdfToWord({ pdfUrl: "<pdfUrl>" });findHelpArticleSearches Adobe's help center (helpx.adobe.com) for an article matching free text.Arguments
Name Type Required querystringYes Returns
HelpArticle[]interface HelpArticle { title: string; product: string; url: string; snippet: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/findHelpArticleCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.findHelpArticle("creative cloud");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.findHelpArticle("creative cloud"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/findHelpArticle \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"creative cloud"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.findHelpArticle("creative cloud");getAgreementStatusReads back one Acrobat Sign agreement the caller sent (the agreementId sendForSignature returns): its state (OUT_FOR_SIGNATURE, SIGNED, …), whether…Arguments
Name Type Required agreementIdstringYes Returns
AgreementStatusinterface AgreementStatus { agreementId: string; name: string; /** Acrobat Sign's own state: OUT_FOR_SIGNATURE, SIGNED, CANCELLED, EXPIRED, … */ status: string; signed: boolean; participants: Array<{ email: string; role: string; order: number | null }>; /** Short-lived download URL of the signed PDF, once status is SIGNED. */ signedDocumentUrl: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/getAgreementStatusCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.getAgreementStatus("CBJCHBCAABAAnobodyHome");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.getAgreementStatus("CBJCHBCAABAAnobodyHome"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/getAgreementStatus \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"agreementId":"CBJCHBCAABAAnobodyHome"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.getAgreementStatus("CBJCHBCAABAAnobodyHome");getCreativeCloudPlansReturns the Individual plans Adobe publishes on www.adobe.com/creativecloud/plans.html.Arguments
None.
Returns
CreativeCloudPlansResultinterface CreativeCloudPlan { name: string; description: string; monthlyPrice: number; regularMonthlyPrice: number | null; promoEndsAt: string | null; promoDuration: string | null; annualPrice: number | null; apps: string[]; } interface CreativeCloudPlansResult { plans: CreativeCloudPlan[]; warnings: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/getCreativeCloudPlansCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.getCreativeCloudPlans();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.getCreativeCloudPlans())curl -X POST https://api.bowmark.ai/v1/sites/adobe/getCreativeCloudPlans \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.getCreativeCloudPlans();getFontReturns one Adobe Fonts family's detail page.Arguments
Name Type Required slugstringYes Returns
FontDetailinterface FontDetail { slug: string; name: string; foundry: string; description: string; weights: Array<{ weight: number; styles: string[]; }>; licensing: { creativeCloud: boolean; foundryLicensed: boolean; }; suggestedPairings: Array<{ slug: string; name: string; }>; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/getFontCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.getFont("open-sans");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.getFont("open-sans"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/getFont \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"slug":"open-sans"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.getFont("open-sans");getStockAssetReturns one Adobe Stock asset's detail.Arguments
Name Type Required idOrUrlstringYes Returns
StockAssetDetailinterface StockAssetDetail { id: string; title: string; mediaType: "photo" | "vector" | "illustration" | "template" | "video" | "3d-asset"; width: number; height: number; keywords: string[]; contributor: string; licensing: Array<{ name: string; price: string }>; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/getStockAssetCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.getStockAsset("<idOrUrl>");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.getStockAsset("<idOrUrl>"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/getStockAsset \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"idOrUrl":"<idOrUrl>"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.getStockAsset("<idOrUrl>");getTemplateReturns one Adobe Express template's detail.Arguments
Name Type Required idstringYes Returns
ExpressTemplateDetailinterface ExpressTemplateDetail { id: string; title: string; category: string; isPremium: boolean; /** The real canvas size where it can be determined; undefined (never fabricated) * otherwise, see the function's own summary. */ width?: number; height?: number; editableElements?: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/getTemplateCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.getTemplate("default-template");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.getTemplate("default-template"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/getTemplate \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"id":"default-template"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.getTemplate("default-template");licenseStockAssetTakes an Adobe Stock asset (id or URL from searchStockAssets) up to the payment step of licensing it and stops there: every licence tier with its…Arguments
Name Type Required idOrUrlstringYes licensestringNo Returns
StockLicenseQuoteinterface StockAssetDetail { id: string; title: string; mediaType: "photo" | "vector" | "illustration" | "template" | "video" | "3d-asset"; width: number; height: number; keywords: string[]; contributor: string; licensing: Array<{ name: string; price: string }>; } interface StockLicenseQuote { id: string; title: string; mediaType: StockAssetDetail["mediaType"]; url: string; /** Every licence tier the page offers, with the price shown logged out. */ licenses: Array<{ name: string; price: string }>; /** The tier asked for (or the first offered), with what the caller would be charged. */ selected: { name: string; price: string } | null; /** Always false: Bowmark stops at the payment step and never charges. */ charged: false; warnings: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/licenseStockAssetCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.licenseStockAsset("<idOrUrl>");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.licenseStockAsset("<idOrUrl>"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/licenseStockAsset \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"idOrUrl":"<idOrUrl>"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.licenseStockAsset("<idOrUrl>");mergePdfsRuns Adobe Acrobat's free online "Merge PDF" tool on two or more caller-supplied PDFs and returns one combined PDF in the order given, the way a…Arguments
Name Type Required pdfUrlsstring[]Two to twelve public https PDF URLs (each up to 5 MB), merged in this order.Yes namestringFile name of the merged PDF; defaults to "merged.pdf".No Returns
AcrobatPdfResultinterface AcrobatPdfResult { name: string; mimetype: string; /** A signed Adobe link to the result; short-lived, so fetch or save it promptly. */ downloadUrl: string; /** Size of the result in bytes. */ bytes: number; /** Combined size of the input PDFs in bytes. */ inputBytes: number; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/mergePdfsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.mergePdfs({ pdfUrls: [] });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.mergePdfs({"pdfUrls": []}))curl -X POST https://api.bowmark.ai/v1/sites/adobe/mergePdfs \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"pdfUrls":[]}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.mergePdfs({ pdfUrls: [] });searchFontsSearches Adobe Fonts (fonts.adobe.com) by family name, returning each matching font family's slug and name (foundry comes from getFont).Arguments
Name Type Required querystringYes Returns
Font[]interface Font { slug: string; name: string; foundry: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/searchFontsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.searchFonts("Helvetica");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.searchFonts("Helvetica"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/searchFonts \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"Helvetica"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.searchFonts("Helvetica");searchStockAssetsSearches Adobe Stock (stock.adobe.com) for photos, vectors, illustrations, templates and video the way the site's own search box does, returning each…Arguments
Name Type Required querystringYes Returns
StockAsset[]interface StockAsset { id: string; title: string; thumbnail: string; /** The asset's canonical stock.adobe.com detail URL, pass it to getStockAsset; * a video, template or 3D asset cannot be opened from its bare id. */ url: string; mediaType: "photo" | "vector" | "illustration" | "template" | "video" | "3d-asset"; orientation?: "landscape" | "portrait" | "square"; licensingPriceTier?: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/searchStockAssetsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.searchStockAssets("drone footage ocean waves");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.searchStockAssets("drone footage ocean waves"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/searchStockAssets \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"drone footage ocean waves"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.searchStockAssets("drone footage ocean waves");searchTemplatesSearches Adobe Express (new.express.adobe.com) templates by free text, optionally narrowed to one format/category, returning each template's id…Arguments
Name Type Required querystringYes options.formatstringOne of the site's own format/category names a result's category field returns, e.g. "flyer", "instagram-story", "card-greeting".No Returns
ExpressTemplate[]interface ExpressTemplate { id: string; title: string; thumbnail: string; category: string; isPremium: boolean; /** Set only when this row may be a generic fallback recommendation rather than a * real match for the search query. */ warning?: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/searchTemplatesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.searchTemplates("flyer");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.searchTemplates("flyer"))curl -X POST https://api.bowmark.ai/v1/sites/adobe/searchTemplates \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"flyer"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.searchTemplates("flyer");
saveAssetToCreativeCloudSaves an image (public https URL, under 5 MB) into the signed-in caller's own Creative Cloud Libraries.Arguments
Name Type Required assetUrlstringA public https URL of the image (PNG, JPEG, GIF, SVG, WebP), under 5 MB.Yes namestringElement name in the library; defaults to the URL's file name.No libraryIdstringA library URN (urn:aaid:sc:…) of the caller's own; defaults to "My Library", made if the account has none.No Returns
SavedCreativeCloudAssetinterface SavedCreativeCloudAsset { libraryId: string; /** Empty when the caller named the library by id. */ libraryName: string; elementId: string; name: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/saveAssetToCreativeCloudCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.saveAssetToCreativeCloud({ assetUrl: "<assetUrl>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.saveAssetToCreativeCloud({"assetUrl": "<assetUrl>"}))curl -X POST https://api.bowmark.ai/v1/sites/adobe/saveAssetToCreativeCloud \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"assetUrl":"<assetUrl>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.saveAssetToCreativeCloud({ assetUrl: "<assetUrl>" });sendForSignatureSends a PDF out for e-signature from the caller's own Adobe account through Acrobat Sign: a public PDF URL and one or more signer emails, signed in…Arguments
Name Type Required documentUrlstringA public https URL of the PDF to send; Acrobat Sign fetches it itself.Yes signersstring[]Signer email addresses, signed in this order.Yes namestringThe agreement's name; defaults to the file name in documentUrl.No messagestringOptional note shown to every signer.No Returns
SignatureRequestinterface SignatureRequest { agreementId: string; name: string; signers: string[]; status: "OUT_FOR_SIGNATURE"; /** Where the caller manages it in Acrobat Sign's web app, when Sign said. */ webAccessPoint: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/adobe/sendForSignatureCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.adobe.sendForSignature({ documentUrl: "<documentUrl>", signers: [] });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.adobe.sendForSignature({"documentUrl": "<documentUrl>", "signers": []}))curl -X POST https://api.bowmark.ai/v1/sites/adobe/sendForSignature \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"documentUrl":"<documentUrl>","signers":[]}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.adobe.sendForSignature({ documentUrl: "<documentUrl>", signers: [] });
Use it from code
- Node
npm i @bowmark/web - Python
pip install bowmark-web - HTTP
POST /v1/sites/adobe/<function> - OpenAPI
/v1/sites/adobe/openapi.json