{
  "title": "Screenshots",
  "description": "Capture a screen, window, or browser page for use as model input.",
  "slug": "screenshots",
  "locale": "en",
  "seoTitle": "Screenshots — Can My Agent Use",
  "socialTitle": "Screenshots",
  "socialDescription": "Compare screenshot capture scope by exact product surface.",
  "llmSummary": "Screenshot support captures a screen, window, or browser page for use as model input. Browser-only and arbitrary desktop capture are distinct qualifiers.",
  "audience": "Engineers comparing chat, desktop, and CLI agent harnesses.",
  "contentKind": "feature",
  "status": "published",
  "tags": [
    "perception"
  ],
  "updated": "2026-08-28T00:00:00.000Z",
  "published": "2026-08-28T00:00:00.000Z",
  "category": "perception",
  "summary": "Capture a screen, window, or browser page for use as model input.",
  "specLabel": "Common product term",
  "aliases": [],
  "capabilityKind": "atomic",
  "related": [],
  "relations": [],
  "highlight": false,
  "notes": [
    {
      "id": 1,
      "text": "Evidence checked 2026-08-28: the first-party OpenWork Browser can take screenshots during browser-control tasks; arbitrary desktop capture is not documented."
    },
    {
      "id": 2,
      "text": "Evidence checked 2026-08-28: VS Code Copilot Chat can capture the current integrated-browser viewport and attach it as image context. Arbitrary desktop capture is not established by the current page."
    },
    {
      "id": 3,
      "text": "Evidence checked 2026-08-28: Gemini CLI's optional browser agent can capture and analyze browser screenshots when a visual model is configured. The visual agent requires API-key or Vertex AI authentication and is unavailable with Sign in with Google."
    }
  ],
  "issues": [],
  "resources": [
    {
      "title": "Methodology",
      "href": "/methodology",
      "kind": "note"
    },
    {
      "id": "openwork-screenshots",
      "title": "OpenWork — Control the browser",
      "href": "https://openworklabs.com/docs/start-here/do-work-with-it/control-the-browser",
      "kind": "docs",
      "publisher": "OpenWork",
      "evidenceType": "documented",
      "reviewedAt": "2026-08-28"
    },
    {
      "id": "vscode-chat-context",
      "title": "Visual Studio Code — Add context to chat",
      "href": "https://code.visualstudio.com/docs/chat/copilot-chat-context",
      "kind": "docs",
      "publisher": "Microsoft",
      "evidenceType": "documented",
      "reviewedAt": "2026-08-28",
      "locator": "Add browser context"
    },
    {
      "id": "gemini-cli-browser-agent",
      "title": "Gemini CLI — Subagents",
      "href": "https://geminicli.com/docs/core/subagents/",
      "kind": "docs",
      "publisher": "Google",
      "evidenceType": "documented",
      "reviewedAt": "2026-08-28",
      "locator": "Browser agent; visual agent"
    }
  ],
  "support": [
    {
      "harness": "openwork-desktop",
      "versions": [
        {
          "track": "current",
          "status": "partial",
          "noteIds": [
            1
          ],
          "target": {
            "kind": "dated-documentation",
            "revision": "current OpenWork Browser documentation",
            "observedAt": "2026-08-28"
          },
          "environmentProfile": "local-default",
          "qualifiers": [
            {
              "type": "runtime",
              "value": "browser screenshots only; arbitrary desktop capture is not established"
            }
          ],
          "evidence": [
            {
              "resourceId": "openwork-screenshots",
              "type": "documented",
              "observedAt": "2026-08-28"
            }
          ]
        }
      ]
    },
    {
      "harness": "vscode-copilot",
      "versions": [
        {
          "track": "current",
          "status": "partial",
          "noteIds": [
            2
          ],
          "target": {
            "kind": "dated-documentation",
            "revision": "2026-08-26 VS Code chat-context documentation",
            "observedAt": "2026-08-28"
          },
          "environmentProfile": "local-default",
          "qualifiers": [
            {
              "type": "runtime",
              "value": "the integrated browser can capture its current viewport and attach it as an image to Chat"
            },
            {
              "type": "runtime",
              "value": "arbitrary desktop, application-window, and multi-monitor capture is not established by the current reviewed page"
            }
          ],
          "evidence": [
            {
              "resourceId": "vscode-chat-context",
              "type": "documented",
              "observedAt": "2026-08-28"
            }
          ]
        }
      ]
    },
    {
      "harness": "gemini-cli",
      "versions": [
        {
          "track": "current",
          "status": "partial",
          "noteIds": [
            3
          ],
          "target": {
            "kind": "dated-documentation",
            "revision": "current Gemini CLI browser-agent documentation",
            "observedAt": "2026-08-28"
          },
          "environmentProfile": "local-default",
          "qualifiers": [
            {
              "type": "runtime",
              "value": "the optional browser agent gains an analyze_screenshot tool that captures a browser screenshot and sends it to a configured vision model"
            },
            {
              "type": "plan",
              "value": "requires API-key or Vertex AI authentication and is unavailable with Sign in with Google"
            },
            {
              "type": "runtime",
              "value": "browser-only capture does not establish arbitrary desktop, application-window, or multi-monitor screenshot support"
            }
          ],
          "evidence": [
            {
              "resourceId": "gemini-cli-browser-agent",
              "type": "documented",
              "observedAt": "2026-08-28"
            }
          ]
        }
      ]
    }
  ],
  "html": "/features/screenshots",
  "markdown": "/features/screenshots.md",
  "json": "/api/v1/features/screenshots.json",
  "body": "Capture the current screen or window as context. OpenWork's sourced cell is deliberately partial because its documented screenshot path is limited to the first-party browser.\n\nThe matrix below lists published **web**, **desktop**, and **CLI** surfaces. Unreviewed cells remain **unknown** until a dated note and public source support a more specific status.\n\nUse the Markdown and JSON twins if you are an agent reading this site."
}