Add browse get markdown command

[derekmeegan]

Summary

• Adds browse get markdown [selector] to convert page HTML to clean markdown
• Defaults to body content when no selector given, accepts optional CSS/XPath/ref selector
• Uses node-html-markdown for quality conversion (links, tables, code blocks preserved)
• Useful for agents that need readable page content without HTML noise

Usage

browse get markdown                # full page body as markdown
browse get markdown .article       # specific element
browse get markdown @0-5           # ref from snapshot

Test results

Test	Local	Remote (Browserbase)
get markdown (body default)	HN full page markdown	HN full page markdown
get markdown .titleline (selector)	Clean link with title	Clean link with title

Test plan

• Test locally with no selector (full body)
• Test locally with CSS selector
• Test on remote Browserbase session (no selector)
• Test on remote Browserbase session (with selector)

🤖 Generated with Claude Code

[changeset-bot]

🦋 Changeset detected

Latest commit: f3d5b5d

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 0 packages

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[cubic-dev-ai]

No issues found across 4 files

Confidence score: 5/5

• Automated review surfaced no issues in the provided summaries.
• No files require special attention.

Architecture diagram

sequenceDiagram
    participant User
    participant CLI as CLI (Commander)
    participant Exec as executeCommand()
    participant SH as Stagehand/Page
    participant Browser as Browser (Local/Remote)
    participant MD as node-html-markdown

    Note over User,MD: NEW: Markdown Extraction Flow

    User->>CLI: browse get markdown [selector]
    CLI->>Exec: action(what="markdown", selector)
    
    alt Selector provided
        Exec->>Exec: resolveSelector(selector)
    else No selector
        Exec->>Exec: CHANGED: Default to "body"
    end

    Exec->>SH: deepLocator(target).innerHtml()
    
    Note over SH,Browser: Context: local playwright or remote Browserbase
    SH->>Browser: CDP/Playwright: Get innerHTML
    Browser-->>SH: HTML string
    SH-->>Exec: HTML string

    Exec->>MD: NEW: NodeHtmlMarkdown.translate(html)
    MD-->>Exec: Markdown string
    
    Exec-->>CLI: { markdown: value }
    CLI-->>User: Print Markdown output

Loading

[shrey150]

I think this might do a better as a skill (/markdown) given that it's not a primitive wrapping the browser. I'd also prefer we do that while we figure out the ergonomics. @derekmeegan what do you think?

[derekmeegan]

i honestly feel markdown extraction is a scraping primitive in the ai agent era but let's chat more synchronously tomorrow