Skip to content

Add optional keyless Parallel search and page fetching - #2

Open
georgeatparallel wants to merge 1 commit into
OperonAgent:mainfrom
georgeatparallel:parallel-search-5707d608-e807-4ae6-b5ce-e12184cb221b
Open

georgeatparallel wants to merge 1 commit into
OperonAgent:mainfrom
georgeatparallel:parallel-search-5707d608-e807-4ae6-b5ce-e12184cb221b

Conversation

@georgeatparallel

Copy link
Copy Markdown

Summary

Give Operon optional web search and page fetching with Parallel Search MCP. Search uses Fast mode, with no account or API key needed. Ask Operon to use parallel_search for focused excerpts and source links, or parallel_fetch to read specific URLs. Both tools use the existing HTTP MCP client and require no additional dependency.

DuckDuckGo stays available. The new tools follow web-toolset disable controls and normal tool approval prompts, return service errors without switching providers, and leave saved MCP servers untouched.

I work at Parallel.

Related issue

None.

Type of change

  • Bug fix
  • New feature
  • Refactor / cleanup
  • Documentation
  • Tests

Checklist

  • Tests added/updated for the change
  • make test passes locally
  • No secrets, API keys, or personal data committed
  • Updated CHANGELOG.md if user-facing
  • Optional dependencies are lazily imported (core REPL still runs without them)

Notes for reviewers

All 25 new tests and the suite selected by CI pass (2,069 tests). The unrestricted suite has 2,431 passes and 13 desktop/vector-memory failures; clean main reproduces the same 13 failures.

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.

1 participant