Document browser pool memory options

[sjmiller609]

Adds memory: "16GiB" to the browser pool create examples (TypeScript, Python, Go). Verified each example against the production API with SDK v0.113.0.

---

Note

Low Risk
Docs-only updates to code samples; no runtime, API, or security behavior changes.

Overview
Documents the memory pool configuration by adding memory: "16GiB" to the Create a browser pool snippets for TypeScript/JavaScript, Python, and Go in browsers/pools.mdx.

The examples now show how to request 16 GiB per pooled browser alongside existing options like stealth, headless, and timeout_seconds, matching the on-demand browser memory API already described elsewhere.

^{Reviewed by Cursor Bugbot for commit 531db56. Bugbot is set up for automated code reviews on this repo. Configure here.}

[mintlify]

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated
Kernel	🟢 Ready	View Preview	Sep 27, 2026, 10:07 PM

💡 Tip: Enable Automations to automatically generate PRs for you.

[cursor]

Risk: Very Low

This pull request changes one sentence in browsers/pools.mdx. It documents that headful browser pools can set memory: "16GiB" and that those pools need Hypeman hosts in the pool's region.

• Codepaths: documentation only. No application code, API handlers, auth, or config logic changed.
• Blast radius: the Browser Pools docs page. No shared libraries, infrastructure, or production behavior.
• Complexity: a one-line addition, easy to review.

Approving because this is a documentation-only edit with no production logic change. Code owner review is not required on main.

  

_{Sent by Cursor Automation: Assign PR reviewers}

[cursor]

Risk: Very Low (unchanged after the latest push)

Re-checked the current diff against main and against the previously approved commit 4642b35. The pull request still touches only browsers/pools.mdx.

• What changed: three example lines add a memory field ("16GiB" in TypeScript and Python, kernel.BrowserMemoryRequest16GiB in Go). The latest commit also drops the prose sentence about headful pools and Hypeman hosts and shows that setting in the create examples instead.
• Codepaths: documentation only. No application code, API handlers, auth, config, or infrastructure.
• Blast radius: the Browser Pools docs page. No shared libraries or production behavior.
• Complexity: a few lines in code samples, easy to review.

No CODEOWNERS file, so code owner review is not required. Risk did not increase, so the existing approval stays in place and this run does not approve again.

  

_{Sent by Cursor Automation: Assign PR reviewers}

[cursor]

Risk: Very Low (unchanged after the latest push)

Re-checked the current diff against main and against the previously reviewed commit 5f45bda. The pull request still touches only browsers/pools.mdx.

• What changed: the Go create example now uses the string "16GiB" for Memory, matching the TypeScript and Python samples. The full diff is still those three memory lines in the pool create examples.
• Codepaths: documentation only. No application code, API handlers, auth, config, or infrastructure.
• Blast radius: the Browser Pools docs page. No shared libraries or production behavior.
• Complexity: a one-line edit inside an existing code sample.

No CODEOWNERS file, so code owner review is not required. Risk did not increase, so the existing approval stays in place and this run does not approve again.

  

_{Sent by Cursor Automation: Assign PR reviewers}

[cursor]

Risk: Very Low

The current diff against main changes only browsers/pools.mdx (+3 lines). Each create example (TypeScript, Python, Go) adds a memory value of "16GiB".

• Codepaths: documentation only. No application code, API handlers, auth, config, or infrastructure.
• Blast radius: the Browser Pools docs page. No shared libraries or production behavior.
• Complexity: three lines inside existing code samples.

This is product documentation, not an instruction or prompt file. main does not require code owner review (require_code_owner_review is false, and there is no CODEOWNERS file).

An approval from this automation is already on the pull request, and the latest commits do not raise the risk, so this run does not approve again.

  

_{Sent by Cursor Automation: Assign PR reviewers}