Add optional Parallel provider for WebSearch - #174
Open
georgeatparallel wants to merge 2 commits into
Open
georgeatparallel wants to merge 2 commits into
georgeatparallel wants to merge 2 commits into
Conversation
georgeatparallel
marked this pull request as ready for review
September 21, 2026 20:15
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
OpenCursor users can choose Parallel for the built-in WebSearch tool without creating an account or adding an API key. DuckDuckGo stays the default.
The new selector lives under Settings > Agents > Context. It uses the existing stdio MCP client with a pinned
mcp-remotebridge, discoversweb_search, and returns excerpts with source URLs. Search queries go to Parallel when selected; the existing web permission controls still apply. WebFetch keeps its current behavior.For background on the search provider, Artificial Analysis publishes DeepSearchQA results and time per task. Those benchmarks measure provider research performance, not this VS Code integration. The free MCP uses Fast search and returns excerpts the agent can use directly. DuckDuckGo isn't evaluated in that comparison.
Type checks, lint, the production build, and focused routing, settings and MCP tests pass. A hosted call through the WebSearch handler returned useful Node.js documentation excerpts and citations without a Parallel key. The harness supplied a minimal VS Code workspace stub; a full Extension Development Host session hasn't been tested. The full unit suite has two failures that reproduce on unchanged upstream, in file mutation undo and ripgrep tests.
Btw - I'm a devrel engineer at Parallel