chore: delete kv channel

[NathanFlurry]

Description

Please include a summary of the changes and the related issue. Please also include relevant motivation and context.

Type of change

• 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 not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Please describe the tests that you ran to verify your changes.

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

[NathanFlurry]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• chore: remove udb as a dependency of envoy-client #4587 : 2 dependent PRs (#4588 , #4591 )
• chore: revert actor v2 hack #4586
• fix: fix pool metadata protocol version refresh, rivetkit native ws #4585
• feat: US-001 - Define SqliteKv trait in rivetkit-sqlite-native #4584
• feat(rust/engine-runner): minimal rust engine runner #4289
• chore: delete kv channel #4583 👈 (View in Graphite)
• feat(engine): rust runner #4582
• chore: remove non-engine drivers #4574 : 1 other dependent PR (#4581 )
• feat: dynamic actors #4397
• fix: driver test suite & sqlite reliability #4559
• fix: remove serverless token #4572
• fix: misc token fixes #4569
• fix(pb): clean up actor stop decision handling #4566
• fix(envoy): use global instance, add signal handlers #4565
• chore: misc fixes, add pb snapshot test #4561
• fix(cache): make in memory cache global #4560
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[claude]

PR Review: chore: delete kv channel

This is a clean deletion PR removing the entire KV channel subsystem (~2,800 lines). The changes are well-scoped and consistent. A few observations:

Summary Assessment

This is a straightforward removal of a dedicated WebSocket transport (/kv/connect) used by @rivetkit/sqlite-native for VFS KV operations. The deletion is thorough and covers all layers: schema, Rust protocol crate, TypeScript SDK, guard routing, and the runtime router.

Positive Notes

• All related artifacts are removed together (schema, generated types, Rust crate, TypeScript package, workspace entries, routing) with no dangling references left behind.
• CLAUDE.md is updated to remove the now-irrelevant sync instructions, which keeps the guidance accurate.
• The JSDoc comment added to hardCrashActor marking it as test-only is a useful clarification.

Questions / Concerns

What replaces the KV channel?
The PR description doesn't explain what @rivetkit/sqlite-native will use instead of /kv/connect for its VFS KV operations. If the native SQLite VFS is still in use, there should be a follow-up PR (or existing code) routing KV operations through the runner protocol or another channel. If sqlite-native is also being removed, that should be noted. Worth clarifying in the PR description.

Leftover references to @rivetkit/engine-kv-channel-protocol?
The package is removed from rivetkit-typescript/packages/rivetkit/package.json and the pnpm workspace, but it's worth confirming no other packages in the monorepo still depend on @rivetkit/engine-kv-channel-protocol (e.g., other examples or internal packages).

actor-gateway/gateway.ts removal of the /kv/connect skip:
The removed early-return guard that bypassed the gateway for /kv/connect paths is safe to delete since the route no longer exists. Just confirming there's no scenario where external traffic could still hit that path expecting the old behavior.

Minor

• The PR is currently in DRAFT state. Once the replacement transport (or confirmation that sqlite-native is also dropped) is clarified, it looks ready for final review.

Overall this is a clean removal with no obvious issues. The main open question is what replaces the KV transport for native SQLite.