Fix Cursor & Windsurf `mcpServers` json example

[ries9112]

The current .json config shown on the README has a bunch of extra commas, which are not valid:

The correct version would remove the last comma after the last entry of env, as well as the unnecessary ones between curly brackets:

I didn't test the VS Code option, but it also seems incorrect with the commas:

[changeset-bot[bot]]

⚠️ No Changeset found

Latest commit: e874713bb85973919872486e1078fc3fe7d4efdd

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[Crunchyman-ralph]

@ries9112 please resolve conflicts!

[ries9112]

Could you confirm that looks correct with the version? Not sure how that happened, sorry about that.

Also wanted to flag some of the information here doesn't seem accurate: https://github.com/eyaltoledano/claude-task-master/blob/main/docs/models.md

I was surprised on that list to see Gemini 2.0 flash with the highest SWE score at 0.754. Looking at this post: https://deepmind.google/models/gemini/flash/ I see 60.4% for gemini 2.5 flash thinking, and no score for gemini 2.0 flash. Maybe SWE and SWE-bench verified are different, but just wanted to flag that weirdly high score gemini 2.0 flash score. Might send someone not that knowledgeable on models down the wrong path as it's shown as the best one. Sorry for the unrelated comment

[Crunchyman-ralph]

interesting, we update the .md file based on the .json file, feel free to change it and I'll merge your PR

[ries9112]

sorry for missing this earlier @Crunchyman-ralph - updated the json for gemini flash model swe scores to match https://deepmind.google/models/gemini/flash/

[rickydata-indexer]

No rush on this @Crunchyman-ralph, but in my experience issues with the .json config can be a major blocker for a lot of people who aren't super familiar with MCP. Tomorrow I'm guiding my team through the installation of this in a session I lead around AI knowledge sharing, and the .json config tends to always be the trickiest part to get correctly configured for everyone. Ofc I'll just provide them with the corrected version, but in general the less confusion gets created around the .json the better in my experience

[Crunchyman-ralph]

I think you're absolutely right, let me review again and get this merged !

[ries9112]

@Crunchyman-ralph I've merged upstream/next and resolved the conflicts 👍

[ries9112]

Hey @Crunchyman-ralph just checking on this, what's the next step? Just want to make sure we can get it merged before it requires more changes to stay up to date with any other changes being pushed