Document the screenshot tool, reframe computer_action - #473
Draft
masnwilliams wants to merge 1 commit into
Draft
Conversation
The MCP server splits screenshot out of computer_action into its own read-only tool, and points callers at execute_playwright_code as the default way to drive a browser. Add a screenshot reference page covering the region parameter and the coordinate space each capture reports, and rewrite computer_action as the fallback for surfaces a selector can't reach. The action table also gains the write_clipboard and read_clipboard rows it was missing.
Contributor
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Docs for the MCP server change that splits
screenshotout ofcomputer_actioninto its own read-only tool (kernel/kernel-mcp-server#135). Merge after that ships.Changes
reference/mcp-server/tools/screenshot— parameters, theregioncrop, and the coordinate space each capture reports. Full-screen captures return screen coordinates you can pass straight tocomputer_action; cropped captures state the offset you need to add, which is the part callers previously had no way to know.computer_action— rewritten as the fallback for surfaces a selector can't reach (canvas, embedded PDFs, native dialogs, drag).screenshotremoved from the action table, and the "always end with a screenshot" tip replaced with guidance to preferexecute_playwright_codeand to note that coordinate accuracy depends on the calling model's pixel grounding.execute_playwright_code— stated as the primary way to drive a browser, with a pointer toariaSnapshot()for locating elements without reading pixels. The existing tip now links the standalonescreenshottool.docs.json— added the new page and reordered the Tools group soexecute_playwright_code→screenshot→computer_actionreflects what to reach for first.integrations/vercel/eve-extension— the tool list describedcomputer_actionas including screenshot controls, which is no longer true. Updated that line and addedscreenshotto the allow-list example.Out of scope but included
The
computer_actionaction table was missingwrite_clipboardandread_clipboard. Both already existed on the tool; they're added here becauseread_clipboardis one of the terminal actions the surrounding note describes, so the table would be misleading without them.changelog.mdxmentions the old behavior and is left alone as a historical record.Testing
Not rendered locally — the Mintlify CLI wouldn't install in this environment (a dependency's postinstall fails).
docs.jsonparses as valid JSON, and every internal link in the changed files was checked against an existing.mdxpath by hand. Worth a preview check before merging.