@sebastianandreasson/pi-autonomous-agents 0.5.1 → 0.5.2

package/README.md

@@ -232,6 +232,7 @@ Recent versions of the package isolate each run more aggressively:
 - in-progress iteration state persisted before agent work starts
 - stale run locks recovered when the owning PID is gone
 - timeout cleanup kills the full spawned process group, not only the direct child
+- parent-death watchers shut down orphaned supervisor and adapter layers instead of letting them continue under `PPID 1`
 
 That is meant to prevent orphaned timed-out agents or concurrent supervisors from corrupting shared state.
 

package/docs/PI_SUPERVISOR.md

@@ -222,6 +222,7 @@ The built-in adapter mitigates obvious local loops by watching PI RPC tool event
 - a soft `continue` can be sent after inactivity
 - a separate tool-aware watchdog can tolerate long-running `bash` or browser work without treating the turn as dead
 - a hard no-event timeout aborts a wedged turn instead of hanging indefinitely
+- parent-loss shutdown tears down the owned supervisor/adapter/PI child tree instead of allowing orphaned background runs
 
 Important: terminal streaming does not reset the heartbeat by itself. The watchdog keys off PI RPC events and active tool state, not raw shell output.
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@sebastianandreasson/pi-autonomous-agents",
   "private": false,
-  "version": "0.5.1",
+  "version": "0.5.2",
   "type": "module",
   "description": "Portable unattended PI harness for developer/tester/visual-review loops.",
   "license": "MIT",
@@ -16,8 +16,8 @@
     "pi-harness": "./src/cli.mjs"
   },
   "scripts": {
-    "check": "node --check src/cli.mjs && node --check src/pi-clear-history.mjs && node --check src/pi-client.mjs && node --check src/pi-config.mjs && node --check src/pi-flow.mjs && node --check src/pi-heartbeat.mjs && node --check src/pi-history.mjs && node --check src/pi-preflight.mjs && node --check src/pi-prompts.mjs && node --check src/pi-repo.mjs && node --check src/pi-report.mjs && node --check src/pi-rpc-adapter.mjs && node --check src/pi-supervisor.mjs && node --check src/pi-telemetry.mjs && node --check src/pi-visual-once.mjs && node --check src/pi-visual-review.mjs && node --check src/index.mjs && node --check test/pi-heartbeat.test.mjs && node --check test/pi-role-models.test.mjs && node --check test/pi-flow.test.mjs && node --check test/pi-history.test.mjs && node --check test/pi-prompts.test.mjs && node --check test/pi-preflight.test.mjs && node --check test/pi-repo.test.mjs && node --check test/pi-telemetry.test.mjs",
-    "test": "node --test test/pi-heartbeat.test.mjs test/pi-role-models.test.mjs test/pi-flow.test.mjs test/pi-history.test.mjs test/pi-prompts.test.mjs test/pi-preflight.test.mjs test/pi-repo.test.mjs test/pi-telemetry.test.mjs"
+    "check": "node --check src/cli.mjs && node --check src/pi-clear-history.mjs && node --check src/pi-client.mjs && node --check src/pi-config.mjs && node --check src/pi-flow.mjs && node --check src/pi-heartbeat.mjs && node --check src/pi-history.mjs && node --check src/pi-preflight.mjs && node --check src/pi-prompts.mjs && node --check src/pi-repo.mjs && node --check src/pi-report.mjs && node --check src/pi-rpc-adapter.mjs && node --check src/pi-supervisor.mjs && node --check src/pi-telemetry.mjs && node --check src/pi-visual-once.mjs && node --check src/pi-visual-review.mjs && node --check src/index.mjs && node --check test/pi-heartbeat.test.mjs && node --check test/pi-lifecycle.test.mjs && node --check test/pi-role-models.test.mjs && node --check test/pi-flow.test.mjs && node --check test/pi-history.test.mjs && node --check test/pi-prompts.test.mjs && node --check test/pi-preflight.test.mjs && node --check test/pi-repo.test.mjs && node --check test/pi-telemetry.test.mjs && node --check test/fixtures/fake-pi.mjs",
+    "test": "node --test test/pi-heartbeat.test.mjs test/pi-lifecycle.test.mjs test/pi-role-models.test.mjs test/pi-flow.test.mjs test/pi-history.test.mjs test/pi-prompts.test.mjs test/pi-preflight.test.mjs test/pi-repo.test.mjs test/pi-telemetry.test.mjs"
   },
   "files": [
     "src",

package/src/cli.mjs

@@ -4,6 +4,11 @@ import path from 'node:path'
 import { spawn } from 'node:child_process'
 import process from 'node:process'
 import { fileURLToPath } from 'node:url'
+import {
+  registerOwnedChildProcess,
+  signalChildProcess,
+  watchParentProcess,
+} from './pi-repo.mjs'
 
 const scriptDir = path.dirname(fileURLToPath(import.meta.url))
 
@@ -36,10 +41,53 @@ function main() {
     env: process.env,
     stdio: 'inherit',
   })
+  registerOwnedChildProcess(child)
+
+  let shuttingDown = false
+  let forceKillTimer = null
+  const stopWatchingParent = watchParentProcess(() => {
+    shutdown({
+      signal: 'SIGTERM',
+      exitCode: 1,
+    })
+  })
+
+  function shutdown({
+    signal,
+    exitCode,
+  }) {
+    if (shuttingDown) {
+      return
+    }
+
+    shuttingDown = true
+    stopWatchingParent()
+    signalChildProcess(child.pid, signal)
+    forceKillTimer = setTimeout(() => {
+      signalChildProcess(child.pid, 'SIGKILL')
+    }, 1000)
+    if (typeof forceKillTimer.unref === 'function') {
+      forceKillTimer.unref()
+    }
+    process.exitCode = exitCode
+  }
+
+  for (const signal of ['SIGINT', 'SIGTERM', 'SIGHUP']) {
+    process.on(signal, () => {
+      shutdown({
+        signal,
+        exitCode: 128,
+      })
+    })
+  }
 
   child.on('exit', (code, signal) => {
+    stopWatchingParent()
+    if (forceKillTimer) {
+      clearTimeout(forceKillTimer)
+    }
     if (signal) {
-      process.kill(process.pid, signal)
+      process.exitCode = 128
       return
     }
     process.exitCode = code ?? 1

package/src/pi-repo.mjs

@@ -114,6 +114,57 @@ export function isProcessRunning(pid) {
   }
 }
 
+const ownedChildren = new Map()
+
+export function registerOwnedChildProcess(child, options = {}) {
+  const pid = normalizePid(child?.pid)
+  if (pid <= 0) {
+    return () => {}
+  }
+
+  const entry = {
+    useProcessGroup: options.useProcessGroup === true && process.platform !== 'win32',
+  }
+  ownedChildren.set(pid, entry)
+
+  const unregister = () => {
+    ownedChildren.delete(pid)
+  }
+
+  if (typeof child?.once === 'function') {
+    child.once('exit', unregister)
+    child.once('close', unregister)
+  }
+
+  return unregister
+}
+
+export function signalChildProcess(pid, signal, options = {}) {
+  const normalizedPid = normalizePid(pid)
+  if (normalizedPid <= 0) {
+    return false
+  }
+
+  try {
+    if (options.useProcessGroup === true && process.platform !== 'win32') {
+      process.kill(-normalizedPid, signal)
+    } else {
+      process.kill(normalizedPid, signal)
+    }
+    return true
+  } catch {
+    return false
+  }
+}
+
+export function signalOwnedChildProcesses(signal) {
+  let handled = false
+  for (const [pid, entry] of [...ownedChildren.entries()]) {
+    handled = signalChildProcess(pid, signal, entry) || handled
+  }
+  return handled
+}
+
 export async function readJsonFile(filePath, fallback = null) {
   try {
     const raw = await fs.readFile(filePath, 'utf8')
@@ -211,20 +262,45 @@ export async function releaseRunLock(lockFile, runId) {
 }
 
 export function signalProcessTree(pid, signal) {
-  const normalizedPid = normalizePid(pid)
-  if (normalizedPid <= 0) {
-    return false
+  return signalChildProcess(pid, signal, { useProcessGroup: true })
+}
+
+export function watchParentProcess(onParentExit, options = {}) {
+  const expectedParentPid = normalizePid(options.parentPid ?? process.ppid)
+  if (expectedParentPid <= 0 || typeof onParentExit !== 'function') {
+    return () => {}
   }
 
-  try {
-    if (process.platform !== 'win32') {
-      process.kill(-normalizedPid, signal)
-    } else {
-      process.kill(normalizedPid, signal)
+  let active = true
+  const intervalMs = Number.isFinite(Number(options.intervalMs))
+    ? Math.max(100, Number(options.intervalMs))
+    : 1000
+
+  const interval = setInterval(() => {
+    if (!active) {
+      return
     }
-    return true
-  } catch {
-    return false
+
+    const currentParentPid = normalizePid(process.ppid)
+    if (currentParentPid === expectedParentPid && currentParentPid > 1) {
+      return
+    }
+
+    active = false
+    clearInterval(interval)
+    onParentExit({
+      expectedParentPid,
+      currentParentPid,
+    })
+  }, intervalMs)
+
+  if (typeof interval.unref === 'function') {
+    interval.unref()
+  }
+
+  return () => {
+    active = false
+    clearInterval(interval)
   }
 }
 
@@ -474,6 +550,9 @@ export async function runShellCommand({
       detached: process.platform !== 'win32',
       stdio: ['pipe', 'pipe', 'pipe'],
     })
+    const unregisterChild = registerOwnedChildProcess(child, {
+      useProcessGroup: process.platform !== 'win32',
+    })
 
     let stdout = ''
     let stderr = ''
@@ -506,6 +585,7 @@ export async function runShellCommand({
     })
 
     child.on('error', (error) => {
+      unregisterChild()
       if (killTimer) {
         clearTimeout(killTimer)
       }
@@ -524,6 +604,7 @@ export async function runShellCommand({
     })
 
     child.on('close', (code) => {
+      unregisterChild()
       if (killTimer) {
         clearTimeout(killTimer)
       }

package/src/pi-rpc-adapter.mjs

@@ -10,7 +10,12 @@ import {
   getHeartbeatDecision,
   resolveHeartbeatConfig,
 } from './pi-heartbeat.mjs'
-import { signalProcessTree } from './pi-repo.mjs'
+import {
+  registerOwnedChildProcess,
+  signalOwnedChildProcesses,
+  signalProcessTree,
+  watchParentProcess,
+} from './pi-repo.mjs'
 
 function createJsonlReader(stream, onLine) {
   const rl = createInterface({ input: stream })
@@ -155,6 +160,9 @@ async function run() {
     detached: process.platform !== 'win32',
     stdio: ['pipe', 'pipe', 'pipe'],
   })
+  registerOwnedChildProcess(child, {
+    useProcessGroup: process.platform !== 'win32',
+  })
 
   let stderr = ''
   const events = []
@@ -197,6 +205,36 @@ async function run() {
   let continueAttempted = false
   let continueAccepted = false
   let continueRejected = false
+  let shutdownRequested = false
+  let shutdownReason = ''
+  let shutdownEscalationTimer = null
+
+  const requestShutdown = (reason, signal = 'SIGTERM') => {
+    if (shutdownRequested) {
+      return
+    }
+
+    shutdownRequested = true
+    shutdownReason = String(reason ?? 'adapter_shutdown')
+    closeAssistantLine()
+    signalOwnedChildProcesses(signal)
+    shutdownEscalationTimer = setTimeout(() => {
+      signalOwnedChildProcesses('SIGKILL')
+    }, 1000)
+    if (typeof shutdownEscalationTimer.unref === 'function') {
+      shutdownEscalationTimer.unref()
+    }
+  }
+
+  const stopWatchingParent = watchParentProcess(() => {
+    requestShutdown('parent_exit', 'SIGTERM')
+  })
+
+  for (const signal of ['SIGINT', 'SIGTERM', 'SIGHUP']) {
+    process.on(signal, () => {
+      requestShutdown(signal, signal)
+    })
+  }
 
   const writeLive = (text) => {
     if (!streamTerminal) {
@@ -460,6 +498,26 @@ async function run() {
 
     await waitForAgentEnd()
 
+    if (shutdownRequested) {
+      console.log(JSON.stringify({
+        sessionId: request.sessionId ?? '',
+        sessionFile: request.sessionFile ?? '',
+        status: 'failed',
+        output: '',
+        notes: shutdownReason,
+        role: '',
+        model: requestedModel,
+        toolCalls: 0,
+        toolErrors: 0,
+        messageUpdates: 0,
+        stopReason: '',
+        loopDetected: false,
+        loopSignature: '',
+        terminalReason: 'adapter_shutdown',
+      }))
+      return
+    }
+
     if (heartbeatTimedOut) {
       const toolCalls = events.filter((event) => event.type === 'tool_execution_start').length
       const toolErrors = events.filter((event) => event.type === 'tool_execution_end' && event.isError).length
@@ -571,9 +629,13 @@ async function run() {
       terminalReason,
     }))
   } finally {
+    stopWatchingParent()
     if (heartbeatInterval) {
       clearInterval(heartbeatInterval)
     }
+    if (shutdownEscalationTimer) {
+      clearTimeout(shutdownEscalationTimer)
+    }
     stopReading()
     for (const current of pending.values()) {
       current.reject(new Error('RPC adapter shutting down'))

package/src/pi-supervisor.mjs

@@ -30,11 +30,13 @@ import {
   releaseRunLock,
   runVerification,
   runShellCommand,
+  signalOwnedChildProcesses,
   stageFiles,
   unstageFiles,
   updateRunLock,
   runVisualCapture,
   timestamp,
+  watchParentProcess,
   writeChangedFiles,
   writeSessionId,
   writeState,
@@ -49,13 +51,30 @@ import {
 import { runStartupPreflight } from './pi-preflight.mjs'
 
 let stopRequested = false
+let shutdownEscalationTimer = null
 
-process.on('SIGINT', () => {
+function requestStop() {
   stopRequested = true
-})
+  signalOwnedChildProcesses('SIGTERM')
+
+  if (!shutdownEscalationTimer) {
+    shutdownEscalationTimer = setTimeout(() => {
+      signalOwnedChildProcesses('SIGKILL')
+    }, 1000)
+    if (typeof shutdownEscalationTimer.unref === 'function') {
+      shutdownEscalationTimer.unref()
+    }
+  }
+}
 
-process.on('SIGTERM', () => {
-  stopRequested = true
+for (const signal of ['SIGINT', 'SIGTERM', 'SIGHUP']) {
+  process.on(signal, () => {
+    requestStop()
+  })
+}
+
+const stopWatchingParent = watchParentProcess(() => {
+  requestStop()
 })
 
 function sleep(seconds) {
@@ -1728,6 +1747,10 @@ async function main() {
       await appendLog(config.logFile, 'Stop requested by signal')
     }
   } finally {
+    stopWatchingParent()
+    if (shutdownEscalationTimer) {
+      clearTimeout(shutdownEscalationTimer)
+    }
     await updateRunOwnership(config, {
       status: stopRequested ? 'stopped' : 'finished',
       heartbeatAt: timestamp(),