Skip to content

Proposal: keyless Parallel Search MCP client example #1332

Description

@georgeatparallel

Would a standalone Parallel Search MCP client example fit in examples/clients? It would demonstrate connecting to a hosted Streamable HTTP server, discovering its tools, and printing web search and page fetch results without running a local server or configuring a Parallel API key.

I propose adding clients_parallel_search alongside the existing HTTP client example, with a query argument and an optional URL to fetch. It would use the existing rmcp transport and list_all_tools / call_tool APIs against https://search.parallel.ai/mcp, with run instructions in the client examples README. Existing examples and SDK defaults would stay unchanged.

Planned validation: run the documented Cargo command in a clean environment without credentials, verify discovered tool schemas, and exercise search and fetch through the example with useful output. Parallel Search MCP supports anonymous use at lower rate limits.

I work at Parallel. I am raising this before a PR to follow the discuss-first contribution guide. Would this be a useful example to include?

Activity

  1. added
    enhancementNew feature or request
    P3Low: nice-to-have, long-term, or exploratory
    T-examplesExample code changes
    ready for workIssue is well-defined and ready to be picked up
    on Oct 9, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    P3Low: nice-to-have, long-term, or exploratoryT-documentationDocumentation improvementsT-examplesExample code changesenhancementNew feature or requestready for workIssue is well-defined and ready to be picked up

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions