Docs: add Parallel Search MCP setup example - #479
Open
georgeatparallel wants to merge 1 commit into
Open
Conversation
georgeatparallel
marked this pull request as ready for review
September 5, 2026 01:25
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds a Parallel Search example to the existing MCP setup section, with the exact Stdio form values and removal steps. Users can connect web search and page extraction without a Parallel account or API key.
Opcode's MCP manager reads user-configured servers. I didn't find a named search provider configured by the repository or a search-provider preset covered by Artificial Analysis. Opcode also displays Claude Code's built-in WebSearch tool, but that isn't a provider result on this leaderboard. There isn't a direct comparison or a measured case for replacing a default here.
As general background, Artificial Analysis's Search API benchmark gives Parallel advanced 80.6/100 on DeepSearchQA F1, which measures research questions with list answers. Parallel fast records 19.2 seconds per task, combining measured search time with derived answer-model time. I pulled those figures directly from the page on September 4, 2026; its latest data is labeled August 31.
Those are two separate API configurations in Artificial Analysis's fixed test harness. They give users a reason to evaluate Parallel, but they don't measure this free MCP recipe, Opcode, or Claude Code's built-in search. This PR adds an optional setup example.
This is a README-only change. The recipe uses
mcp-remotefor Streamable HTTP and explains the all-project user scope, existing tool permissions, which inputs go to Parallel, and free-access rate limits.Validation: saved the documented command through the prebuilt Claude Code CLI (2.1.210), which opcode's MCP commands use, in an isolated configuration directory.
claude mcp get parallel-searchreported Connected. An MCP SDK client loaded the saved command and arguments, discoveredweb_searchandweb_fetch, and successfully called both without Parallel credentials throughmcp-remote0.8.3 on Node.js 24.18.0. Public links andgit diff --checkpassed. The Tauri GUI, repository build, and Rust suite were not run.I work at Parallel, which operates this service.