docs: add Parallel Search MCP setup example - #2557
Open
George Pickett (georgeatparallel) wants to merge 1 commit into
Open
docs: add Parallel Search MCP setup example#2557George Pickett (georgeatparallel) wants to merge 1 commit into
George Pickett (georgeatparallel) wants to merge 1 commit into
Conversation
Contributor
|
Thank you for your contribution! Before we can merge this PR, we need you to sign our Contributor License Agreement. To sign the CLA, please add a comment to this PR with the following text: You only need to sign once. After signing, this check will pass automatically. Troubleshooting
|
George Pickett (georgeatparallel)
marked this pull request as ready for review
September 5, 2026 03:15
Contributor
|
PR author is not in the allowed authors list. |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This adds a Parallel Search example to the custom MCP setup guide, so users can get public web search and page extraction without a Parallel account, API key, or local proxy. It also fixes the SSE-only wording: BrowserOS already detects Streamable HTTP.
Brave is the only search provider in the managed connector catalog, and BrowserOS includes that catalog when opening its managed MCP session. Authentication is still separate. That's the comparison I focused on.
There's a good reason to try Parallel here. In Artificial Analysis's August 31 results, Parallel advanced scored 80.6 on DeepSearchQA F1, and Parallel fast scored 80.2, versus 62.3 for Brave web and 78.1 for Brave LLM context. The free MCP connection uses fast mode. That mode also averaged 19.2 seconds per task, versus Brave web's 35.9 and Brave LLM context's 24.1.
For research-heavy agent tasks, I'd consider testing Parallel as the default search option in place of Brave. Brave LLM context does score higher than Parallel fast on the overall index, 74.6 versus 73.1. These are API benchmark results, with task time including model and search time; BrowserOS's own agent still needs a comparison, and its Klavis connector doesn't pin which Brave API variant it uses.
This PR only changes docs. It leaves defaults and permissions alone, and explains what goes to Parallel, the free-access limits, and how to remove the connection.
Checked the example through BrowserOS's
BrowserContextSchema,buildMcpServerSpecs, andcreateMcpClients, using its locked@ai-sdk/mcp2.0.40 dependency with credentials absent. Transport detection, discovery, oneweb_search, oneweb_fetch, and cleanup passed. MDX syntax andgit diff --checkpassed too. I didn't run the browser UI, a repository build, or the full test suite.I work at Parallel, which operates this service.