Replace httpx_client_factory with direct httpx_client parameter for better flexibility

[ruhz3]

Added support for the verify option to the create_mcp_http_client function and related client factories, allowing users to control SSL certificate verification when creating an httpx AsyncClient.

Motivation and Context

Some users need to disable SSL verification (e.g., for testing or internal endpoints), or use a custom CA bundle. This change makes it possible to pass verify=True, verify=False, or a custom ssl.SSLContext when creating MCP HTTP clients.

How Has This Been Tested?

Tested by creating AsyncClient instances with different verify values (True, False, and custom context) and confirming correct behavior for secure and insecure endpoints. Existing unit tests for HTTP client creation were also run.

Breaking Changes

No breaking changes. The new verify parameter is optional and defaults to the previous behavior.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Documentation
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have added or updated documentation as needed

Additional context

This change improves flexibility for users working in environments with custom or self-signed certificates, or those who need to disable SSL verification for

resolves #870
This PR does not fully resolve the linked issue, but it applies the necessary modifications to the current repository. Further functional modifications following this PR are needed in the fastmcp repository.

[Kludex]

We should be passing the client, instead of allowing more and more options.

[ruhz3]

Respecting the original author's intention to avoid exposing AsyncClient directly to users, the FactoryClient was modified to accept kwargs.

[felixweinberger]

IIUC I don't think the intention here was to pass through arbitrary arguments through to the underlying client, but refactoring to remove the factory in the first place and explicitly use the client.

See #1177 for context. I'm going to mark this as pending that change.

[ruhz3]

I’d greatly appreciate any active and constructive reviews.

[felixweinberger]

Hi @ruhz3 thank you for updating this PR - I don't think we'll go with this approach as we're actually intending to remove all the arguments other than the httpx.AsyncClient

We're currently working on this here: #1177

I'm closing this for now, feel free to chime in on the linked PR though if it's missing something you need!