chore(webapp,core): remove the end-of-life v3 (engine V1) execution stack (#4236)

## Summary

v3 (the engine that ran the SDK v3 era, internally
`RunEngineVersion.V1`) is end-of-life. Following the removal of the v3
execution apps
([#4194](#4194)) and
the legacy dev websocket
([#4198](#4198)), this
removes the remaining v3 execution stack from the server.

Clients still on v3 (an old SDK or CLI that has not upgraded) keep
getting a clear "upgrade to v4" response. Triggers, batch triggers,
reschedules, and deploys that resolve to v3 are rejected with a graceful
4xx pointing at the migration guide, never a 5xx, so a stale client
cannot affect server health. Self-hosted instances still running v3
should stay on the 4.5.x release line until they migrate.

## What is removed

- The MarQS queue and its shared/dev queue consumers.
- The v3 socket.io namespaces (coordinator, provider, shared-queue) and
the v3 run lifecycle services (attempt, checkpoint, and batch-resume).
- The graphile-worker background job system; all live jobs already run
on `@trigger.dev/redis-worker`.
- The `DEPRECATE_V3_ENABLED` flag: v3 is now rejected unconditionally,
so the flag is gone.
- Unused v3 exports from `@trigger.dev/core` (the `v3/zodNamespace`
subpath and the legacy socket message catalogs) and the now-dead MarQS
environment variables.

## What stays

The v4 engine is untouched. The graceful v3 rejection boundary stays,
`determineEngineVersion` still detects a v3 project so it can reject it,
and the batch service plus batch-completion worker stay for current
clients. Live queue concurrency limits and metrics now read from the v4
run engine instead of MarQS, and a brand-new dev environment now
defaults to v4.



## Dependency cleanup

Removes webapp dependencies left unused by this change: `seedrandom` and
`semver` (only the removed v3 code used them) plus a set that was
already dead, their orphaned `@types` packages, and two dead files. Adds
a `knip:deps` script and a `knip.json` config so unused dependencies can
be found the same way going forward.

Author: ericallam

.changeset/retire-v3-zod-namespace.md

@@ -0,0 +1,5 @@
+---
+"@trigger.dev/core": patch
+---
+
+Removed the unused `@trigger.dev/core/v3/zodNamespace` export and the legacy v3 socket message schemas. These were only used by the now-retired v3 engine and have no v4 consumers.

.claude/rules/legacy-v3-code.md

@@ -3,31 +3,24 @@ paths:
   - "apps/webapp/app/v3/**"
 ---
 
-# Legacy V1 Engine Code in `app/v3/`
+# v3 (engine V1) has been removed
 
-The `v3/` directory name is misleading - most code here is actively used by the current V2 engine. Only the specific files below are legacy V1-only code.
+The v3 engine (RunEngineVersion `V1`: MarQS queue + Graphile worker) is end-of-life and its execution code has been removed from the webapp. The `app/v3/` directory name is historical: everything under it now serves the current V2 engine (`@internal/run-engine` + `@trigger.dev/redis-worker`).
 
-## V1-Only Files - Never Modify
+There is no `V1` execution path anymore. If you find a `RunEngineVersion` branch, the `V1` arm should only reject or finalize gracefully (for example, mark a historical run cancelled in the DB), never run V1 work. Do not reintroduce MarQS, the graphile worker, or the v3 socket.io namespaces.
 
-- `marqs/` directory (entire MarQS queue system: sharedQueueConsumer, devQueueConsumer, fairDequeuingStrategy, devPubSub)
-- `legacyRunEngineWorker.server.ts` (V1 background job worker)
-- `services/triggerTaskV1.server.ts` (deprecated V1 task triggering)
-- `services/cancelTaskRunV1.server.ts` (deprecated V1 cancellation)
-- `authenticatedSocketConnection.server.ts` (V1 dev WebSocket using DevQueueConsumer)
-- `sharedSocketConnection.ts` (V1 shared queue socket using SharedQueueConsumer)
+## The deprecation boundary (keep this)
 
-## V1/V2 Branching Pattern
+Requests from clients still on v3 (old SDK/CLI) or historical V1 runs must return a clean 4xx, never a 5xx. The boundary lives in:
 
-Some services act as routers that branch on `RunEngineVersion`:
-- `services/cancelTaskRun.server.ts` - calls V1 service or `engine.cancelRun()` for V2
-- `services/batchTriggerV3.server.ts` - uses marqs for V1 path, run-engine for V2
+- `engineDeprecation.server.ts` - the `V3_TRIGGER_DEPRECATION_MESSAGE` / `V3_DEV_DEPRECATION_MESSAGE` / `V3_MIGRATION_URL` upgrade messages.
+- `engineVersion.server.ts` - `determineEngineVersion()` still detects a V1 project/run so callers can reject it.
+- `services/triggerTask.server.ts`, `services/cancelTaskRun.server.ts`, `services/rescheduleTaskRun.server.ts` - the `V1` arm rejects or finalizes gracefully instead of executing.
+- `services/initializeDeployment.server.ts` - the `DEPRECATE_V3_CLI_DEPLOYS_ENABLED`-gated v3 CLI deploy rejection.
+- `handleWebsockets.server.ts` - the legacy `trigger dev` websocket closes with the upgrade message.
 
-When editing these shared services, only modify V2 code paths.
+## V2 modern stack
 
-## V2 Modern Stack
-
-- **Run lifecycle**: `@internal/run-engine` (internal-packages/run-engine)
-- **Background jobs**: `@trigger.dev/redis-worker` (not graphile-worker/zodworker)
-- **Queue operations**: RunQueue inside run-engine (not MarQS)
-- **V2 engine singleton**: `runEngine.server.ts`, `runEngineHandlers.server.ts`
-- **V2 workers**: `commonWorker.server.ts`, `alertsWorker.server.ts`, `batchTriggerWorker.server.ts`
+- **Run lifecycle**: `@internal/run-engine` (`runEngine.server.ts`, `runEngineHandlers.server.ts`)
+- **Background jobs**: `@trigger.dev/redis-worker` (`commonWorker.server.ts`, `alertsWorker.server.ts`, `batchTriggerWorker.server.ts`; `legacyRunEngineWorker.server.ts` still hosts the live batch-completion jobs)
+- **Queue operations**: RunQueue inside run-engine (`runQueue.server.ts`), not MarQS

.server-changes/remove-v3-engine.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: breaking
+---
+
+Removed support for the end-of-life v3 engine. Instances or projects still on v3 must stay on the 4.5.x release line or upgrade to v4; v3 triggers, batch triggers, reschedules, and deploys now return a clear upgrade message instead of running.

AGENTS.md

@@ -138,11 +138,10 @@ User API call -> Webapp routes -> Services -> RunEngine -> Redis Queue -> Superv
 - **internal-packages/redis**: Redis client creation utilities (ioredis)
 - **internal-packages/testcontainers**: Test helpers for Redis/PostgreSQL containers
 - **internal-packages/schedule-engine**: Durable cron scheduling
-- **internal-packages/zod-worker**: Graphile-worker wrapper (DEPRECATED - use redis-worker)
 
-### Legacy V1 Engine Code
+### v3 (engine V1) removed
 
-The `apps/webapp/app/v3/` directory name is misleading - most code there is actively used by V2. Only specific files are V1-only legacy (MarQS queue, triggerTaskV1, cancelTaskRunV1, etc.). See `apps/webapp/CLAUDE.md` for the exact list. When you encounter V1/V2 branching in services, only modify V2 code paths. All new work uses Run Engine 2.0 (`@internal/run-engine`) and redis-worker.
+v3 (engine V1: MarQS + Graphile worker) is end-of-life and its execution code has been removed. The `apps/webapp/app/v3/` directory name is historical - everything there now serves V2 (Run Engine 2.0, `@internal/run-engine` + redis-worker). There is no V1 execution path: a `RunEngineVersion` `V1` branch only rejects or finalizes gracefully so v3 clients get a clean 4xx, never a 5xx. Do not reintroduce V1. See `apps/webapp/CLAUDE.md` and `.claude/rules/legacy-v3-code.md`.
 
 ### Documentation
 

ai/references/repo.md

@@ -23,7 +23,6 @@ This is a pnpm 10.33.2 monorepo that uses turborepo @turbo.json. The following w
 - <root>/internal-packages/run-engine is the `@internal/run-engine` package that is "Run Engine 2.0" and handles moving a run all the way through it's lifecycle
 - <root>/internal-packages/redis is the `@internal/redis` package that exports Redis types and the `createRedisClient` function to unify how we create redis clients in the repo. It's not used everywhere yet, but it's the preferred way to create redis clients from now on.
 - <root>/internal-packages/testcontainers is the `@internal/testcontainers` package that exports a few useful functions for spinning up local testcontainers when writing vitest tests. See our [tests.md](./tests.md) file for more information.
-- <root>/internal-packages/zodworker is the `@internal/zodworker` package that implements a wrapper around graphile-worker that allows us to use zod to validate our background jobs. We are moving away from using graphile-worker as our background job system, replacing it with our own redis-worker package.
 
 ## References
 

apps/supervisor/CLAUDE.md

@@ -7,7 +7,7 @@ Node.js app that manages task execution containers. Receives work from the platf
 - `src/services/` - Core service logic
 - `src/workloadManager/` - Container orchestration abstraction (Docker or Kubernetes)
 - `src/workloadServer/` - HTTP server for workload communication (heartbeats, snapshots)
-- `src/clients/` - Platform communication (webapp/coordinator)
+- `src/clients/` - Platform communication (webapp)
 - `src/env.ts` - Environment configuration
 
 ## Architecture

apps/webapp/CLAUDE.md

@@ -91,24 +91,14 @@ Background job workers use `@trigger.dev/redis-worker`:
 - `app/v3/alertsWorker.server.ts`
 - `app/v3/batchTriggerWorker.server.ts`
 
-Do NOT add new jobs using zodworker/graphile-worker (legacy).
-
 ## Real-time
 
 - Socket.io: `app/v3/handleSocketIo.server.ts`, `app/v3/handleWebsockets.server.ts`
 - Electric SQL: Powers real-time data sync for the dashboard
 
-## Legacy V1 Code
-
-The `app/v3/` directory name is misleading - most code is actively used by V2. Only these specific files are V1-only legacy:
-- `app/v3/marqs/` (old MarQS queue system)
-- `app/v3/legacyRunEngineWorker.server.ts`
-- `app/v3/services/triggerTaskV1.server.ts`
-- `app/v3/services/cancelTaskRunV1.server.ts`
-- `app/v3/authenticatedSocketConnection.server.ts`
-- `app/v3/sharedSocketConnection.ts`
+## v3 (engine V1) removed
 
-Some services (e.g., `cancelTaskRun.server.ts`, `batchTriggerV3.server.ts`) branch on `RunEngineVersion` to support both V1 and V2. When editing these, only modify V2 code paths.
+v3 (engine V1: MarQS + Graphile worker) is end-of-life and its execution code is gone. The `app/v3/` directory name is historical; everything under it now serves V2. There is no V1 execution path: a `RunEngineVersion` `V1` branch (e.g. in `triggerTask.server.ts`, `cancelTaskRun.server.ts`) only rejects/finalizes gracefully so v3 clients get a clean 4xx, never a 5xx. Do not reintroduce V1. See `.claude/rules/legacy-v3-code.md` for the deprecation boundary.
 
 ## Performance: Trigger Hot Path