Skip to content

docs: add Parallel MCP setup example - #742

Open
georgeatparallel wants to merge 2 commits into
browser-use:mainfrom
georgeatparallel:docs/parallel-mcp-setup
Open

docs: add Parallel MCP setup example#742
georgeatparallel wants to merge 2 commits into
browser-use:mainfrom
georgeatparallel:docs/parallel-mcp-setup

Conversation

@georgeatparallel

@georgeatparallel georgeatparallel commented Sep 5, 2026

Copy link
Copy Markdown

Adds an optional Parallel Search MCP recipe to the README so users can upload a working configuration in Agent Settings for public web search and page extraction without a Parallel account or API key.

The example uses the existing stdio client and a pinned mcp-remote bridge. It explains the Node.js requirement, preserving other configured servers, how to disable the tools, and that queries, requested URLs, and supplied context go to Parallel when the agent calls them. Free access is rate limited. No application code, dependencies, or defaults change.

The built-in search option is Google Search in the browser, which this controller inherits without excluding it. I checked Artificial Analysis's Search API benchmark, whose latest data is dated August 31, 2026. It doesn't include Google Search, so those scores can't tell us whether Parallel would improve on this default. This example gives users a way to try that comparison. If Parallel returns better results on your agent's tasks, it could be worth considering as the default in a separate change.

Validation: loaded the exact README JSON through the unchanged upload callback, MCP setup function, controller registration, and selected-action method with langchain_mcp_adapters==0.0.9 and browser-use==0.1.48. Both tools registered; anonymous search returned 10 results and fetch returned one page. This focused check loaded those methods in isolation with prebuilt dependencies; it did not launch the UI or browser or run a repository build. git diff --check passes.

I work at Parallel, which operates this service.


Summary by cubic

Adds an optional Parallel Search MCP setup example to the README so users can enable web search and page extraction in Agent Settings without a Parallel account or API key.

  • The recipe uses the existing stdio client with a pinned mcp-remote bridge and requires Node.js 22+ with npm.
  • Queries, requested URLs, and supplied context are sent to Parallel when the agent calls the tools; free access is rate limited.
  • Uploaded configuration changes take effect on the next task; no WebUI restart is needed.
  • No application code, dependencies, or defaults change.

Written for commit 06ef624. Summary will update on new commits.

Review in cubic

@CLAassistant

CLAassistant commented Sep 5, 2026

Copy link
Copy Markdown

CLA assistant check
All committers have signed the CLA.

@georgeatparallel
georgeatparallel marked this pull request as ready for review September 5, 2026 03:15

@cubic-dev-ai cubic-dev-ai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

All reported issues were addressed across 1 file

Reply with feedback, questions, or to request a fix.

Fix all with cubic | Re-trigger cubic

Comment thread README.md Outdated
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants