fix: resume saved user messages after execution recovery (#13270)

Preserve verified native process-stop evidence and retry saved user messages through normal continuation admission after recovery cleanup. Show the current wait reason and serialize delivery so a saved message starts one fresh turn.

Validated with 410 focused tests, typecheck, build, token gates, all PR CI checks, and Greptile 5/5.

Co-Authored-By: Paperclip <noreply@paperclip.ing>

server/src/services/native-runtime/native-restart-recovery.ts

@@ -1,3 +1,4 @@
+import { recordNativeLocalProcessStop } from "../native-local-process-stop.js";
 import { randomUUID } from "node:crypto";
 import { and, desc, eq, inArray, isNull, lte, or, sql } from "drizzle-orm";
 import type { Db } from "@paperclipai/db";
@@ -909,6 +910,10 @@ export async function claimNativeRestartRecoveries(input: {
         } as const;
       }
 
+      if (claimKind !== "reattach_existing_runner") {
+        await recordNativeLocalProcessStop(tx as unknown as Db, row.run);
+      }
+
       await tx
         .update(heartbeatRuns)
         .set({

server/src/services/native-runtime/native-runner-restart-recovery.integration.test.ts

@@ -1,3 +1,4 @@
+import { hasNativeLocalProcessStop } from "../native-local-process-stop.js";
 import { randomUUID } from "node:crypto";
 import { spawn } from "node:child_process";
 import { existsSync } from "node:fs";
@@ -580,6 +581,10 @@ describeEmbeddedPostgres("native runner restart recovery with real processes", (
       if (!claim || claim.kind !== "resume_dead_runner") {
         throw new Error("Expected dead-runner recovery claim");
       }
+      expect(await hasNativeLocalProcessStop(fixture.db, companyId, fixture.runId)).toBe(true);
+      const [stoppedRun] = await fixture.db.select().from(heartbeatRuns).where(eq(heartbeatRuns.id, fixture.runId));
+      expect(stoppedRun.processPid).toBeNull();
+      expect(stoppedRun.processGroupId).toBeNull();
 
       restored = createRunnerdCodexTransport({
         ...options,

server/src/services/native-runtime/native-session-executor.test.ts

@@ -4119,7 +4119,7 @@ function leaseDb(
             returning: () => Promise<Array<{ runId: string }>>;
           };
           result.returning = () =>
-            Promise.resolve([{ runId: coordinator.runId }]);
+            Promise.resolve([{ runId: coordinator.runId, nextEventSeq: 2 }]);
           return result;
         },
       };
@@ -4162,12 +4162,20 @@ function leaseDb(
       return query;
     },
   });
+  const insert = (table: unknown) => ({
+    values: (values: Record<string, unknown>) => {
+      updates.push({ table, values });
+      return { returning: async () => [values] };
+    },
+  });
   const tx = {
+    insert,
     execute: async () => [],
     select,
     update,
   };
   return {
+    insert,
     select,
     transaction: async (operation: (transaction: Db) => Promise<unknown>) =>
       operation(tx as unknown as Db),
@@ -6463,13 +6471,16 @@ describe("native process ownership", () => {
         highestContiguousSourceSeq: 1,
       };
     });
-    state.createBackend.mockImplementationOnce((_input, options) => ({
-      kind: "test",
-      onSpawn: options.onSpawn,
-    }));
+    const updates: Array<{ table: unknown; values: Record<string, unknown> }> = [];
+    state.createBackend.mockImplementationOnce((_input, options) => {
+      expect(updates).toContainEqual({ table: heartbeatRunEvents, values: expect.objectContaining({
+        eventType: "native.process_start_requested", runId: execution.binding.runId,
+      }) });
+      return { kind: "test", onSpawn: options.onSpawn };
+    });
 
     await executePaperclipNativeSession({
-      db: leaseDb(),
+      db: leaseDb(execution, {}, {}, updates),
       execution,
       runnerInstanceId: "runner",
       onSpawn,

server/src/services/native-runtime/native-session-executor.ts

@@ -1,3 +1,4 @@
+import { PROCESS_START_REQUESTED } from "../native-local-process-stop.js";
 import { remoteLeaseCleanupScope } from "../remote-execution-termination.js";
 import { resolveConnectorAssignments, isConnectorSkill } from "../connector-runtime.js";
 import {
@@ -7598,6 +7599,17 @@ async function executePaperclipNativeSessionWithinScope(
       input.db,
       input.execution.binding,
     );
+    // Invalidate prior stop evidence before a backend can spawn. A crash between
+    // spawn and the PID callback must not make an old receipt authorize a turn.
+    await appendHeartbeatRunEvent(input.db, {
+      companyId: input.execution.binding.companyId,
+      runId: input.execution.binding.runId,
+      agentId: input.execution.binding.agentId,
+      eventType: PROCESS_START_REQUESTED,
+      stream: "system",
+      level: "info",
+      message: "Native execution requested; prior local stop evidence no longer applies.",
+    });
     const runnerdBackend =
       input.useRunnerd && input.backend === undefined
         ? await createRunnerdBackend({