{
  "data": null,
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/niches/cms.upload-media/",
    "json": "https://www.anchorterminal.com/niches/cms.upload-media/index.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/niches/cms.upload-media/index.md",
    "mcp": "https://www.anchorterminal.com/mcp",
    "slim": "https://www.anchorterminal.com/niches/cms.upload-media/index.min.md"
  },
  "markdown": "One company's docs show a full fit.\n\n| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |\n| --- | --- | --- | --- | --- |\n| 1 | 14 | 17 | 14 | 8 |\n\n## What counts as done\n\nAsset stored with alt text, referenced from the named entry or field, rendered on the preview and the editor canvas, with dimensions recorded, without the file's bytes passing through the model's context.\n\nMeasured, for our own tests: in at most three calls.\n\nHow agents ask for it: add these photos to the gallery page; use this image as the featured image; put the generated hero image on the page\n\n## Does the job\n\nThe docs show it meets the test above, and nothing we read says otherwise.\n\n- **Sanity (Sanity US Inc. (with Sanity AS))**: dataset_assets_upload_from_url fetches a public URL server-side, assets_upload_status returns the reference, and patch_documents attaches it with alt text: three calls, no bytes in context.\n\n## Does the job, with a limit\n\nIt does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.\n\n- **BCMS (Becomes tech, LLC)**: request-upload-media-url returns an upload URL, but sending the bytes needs a shell or HTTP step outside MCP, and no alt-text or from-URL parameter is documented.\n- **ButterCMS (ButterCMS LLC)**: Media fields take {url, alt} and the file is copied into the Media Library server-side in the same write, but only behind the paid Write API, and dimensions aren't documented.\n- **Cosmic (Cosmic Inc.)**: cosmic_media_upload takes a URL so bytes stay out of context and media reads give dimensions, but the MCP page doesn't show an alt-text parameter on upload.\n- **DatoCMS (Dato Srl)**: uploads.createFromUrl takes a remote URL with default alt text and runs inside an MCP script, but acc-29 shows the Free plan's upload quota stopping uploads.\n- **Directus (Monospace Inc. (doing business as Directus))**: The files tool's import action fetches a URL server-side, then the item references the file id; Directus files have title and description but no dedicated alt-text field.\n- **dotCMS (dotCMS LLC)**: upload_assets streams files from a local directory to the site and can publish and verify them, bytes staying out of context, but it needs local files, not a URL, and alt text isn't covered.\n- **Ghost (Ghost Foundation Ltd)**: The image endpoint takes a multipart file with a ref, not a remote URL, so an agent needs a shell to send bytes; alt text is set on the post, not on upload.\n- **Hygraph (Hygraph GmbH)**: createAsset with a remote URL through execute_graphql keeps bytes out of context, but the asset is pending until processed and alt text is a custom field, not built in.\n- **Payload (Payload CMS, Inc.)**: REST takes a multipart upload to the upload collection, so bytes go from disk, not context; over MCP every media create fails with MissingFile (acc-21).\n- **Prismic (Prismic.io Inc.)**: upload_asset fetches a URL server-side and returns the CDN URL, and image fields carry alt text; the docs don't say dimensions are recorded or that the result shows on the editor canvas.\n- **Storyblok (Storyblok GmbH)**: upload_asset returns a signed S3 URL and a cURL command, and the docs say finishing needs shell access, so web clients can't complete it.\n- **Strapi (Strapi, Inc.)**: The MCP's media tools write metadata only (media_update_asset sets alternativeText); the file must go up through REST multipart, and no upload-from-URL is documented.\n- **Webflow (Webflow, Inc.)**: create_asset returns a presigned target and the bytes must be POSTed, with alt text set by update_asset; acc-24 shows API-inserted images missing in the Designer and carrying no dimensions.\n- **WordPress (WordPress.org, an open-source project. The WordPress trademark belongs to the WordPress Foundation)**: POST /wp/v2/media takes the file bytes and alt_text, with media_details recording sizes, but there's no upload-from-URL, and acc-35 saw body images dropped silently by a cross-posting tool.\n\n- not this job: Contentstack (Contentstack Inc.): The MCP's 78 cma tools and the cma-extended group have no asset create or upload; only metadata updates and folders.\n- not this job: CookieYes MCP Server (CookieYes Limited): No upload or asset tool in the tool table.\n- not this job: Kontent.ai (Kontent CZ s.r.o.): The MCP's asset tools are get, list and update only; upload needs the Management API's binary file endpoint outside MCP.\n\n## What stops agents finishing\n\n- MCP server has no upload tool or no file parameter (Payload, Strapi)\n- upload needs a presigned S3 step only shell access can do (Storyblok over MCP)\n- base64-only upload forces bytes through context\n- free-tier asset quota\n\n## Around it\n\nBefore: image-generation.create, file-storage.read. After: cms.draft-post, cms.create-page.\n\n## Where this comes from\n\nWeight 14 of 100 in CMS \u0026 website publishing: 8 of 39 accounts (one not on a listing); lowered from 15 for that; upload is the step most often missing from an MCP server\n\nAs data: https://www.anchorterminal.com/api/v1/use-cases/cms.upload-media.json, or the MCP tool get_use_case with id cms.upload-media. Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person.\n",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "page": {
    "breadcrumbs": [
      {
        "name": "Home",
        "url": "https://www.anchorterminal.com/"
      },
      {
        "name": "For agents",
        "url": "https://www.anchorterminal.com/agents/"
      },
      {
        "name": "Niches",
        "url": "https://www.anchorterminal.com/niches/"
      },
      {
        "name": "Upload images and attach them to an entry",
        "url": ""
      }
    ],
    "description": "Upload images and attach them to an entry: 1 companies fully do it, 14 with a limit, 17 carry the tag. Every listing in CMS \u0026 website publishing with the reason and a source.",
    "facts": null,
    "h1": "Upload images and attach them to an entry",
    "image": "https://www.anchorterminal.com/assets/og/default.png",
    "path": "/niches/cms.upload-media/",
    "published": "",
    "section": "agents",
    "title": "Upload images and attach them to an entry | Anchor Terminal",
    "toc": null,
    "updated": "2026-11-15",
    "url": "https://www.anchorterminal.com/niches/cms.upload-media/"
  },
  "tokens": {
    "markdown": 1500,
    "slim": 380
  },
  "version": 1
}
