neoagent 2.1.18-beta.36 → 2.1.18-beta.38

package/README.md

@@ -3,7 +3,7 @@
 <p align="center"><strong>Your agent. Your server. Your rules.</strong></p>
 
 <p align="center">
-  <a href="https://nodejs.org"><img src="https://img.shields.io/badge/Node.js-18+-5fa04e?style=flat-square&logo=node.js&logoColor=white" alt="Node.js"></a>
+  <a href="https://nodejs.org"><img src="https://img.shields.io/badge/Node.js-20+-5fa04e?style=flat-square&logo=node.js&logoColor=white" alt="Node.js"></a>
   <a href="https://sqlite.org"><img src="https://img.shields.io/badge/SQLite-WAL-003b57?style=flat-square&logo=sqlite&logoColor=white" alt="SQLite"></a>
   <a href="https://flutter.dev"><img src="https://img.shields.io/badge/Flutter-web%20%2B%20android-02569B?style=flat-square&logo=flutter&logoColor=white" alt="Flutter"></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-a855f7?style=flat-square" alt="License"></a>

package/docs/getting-started.md

@@ -4,7 +4,7 @@ NeoAgent installs as a Node CLI and runs a self-hosted server with a bundled Flu
 
 ## Requirements
 
-- Node.js 18 or newer.
+- Node.js 20 or newer.
 - A reachable server URL if you want OAuth callbacks, mobile access, or messaging webhooks.
 - At least one hosted AI provider API key, unless you only use local Ollama.
 - Android Studio or a Flutter Android toolchain if you build the Android client yourself.

package/docs/index.md

@@ -1,47 +1,14 @@
 ---
-layout: home
+slug: /
+title: NeoAgent
+sidebar_label: Overview
+---
 
-hero:
-  name: NeoAgent
-  text: Self-hosted proactive AI agent
-  tagline: Run your own server, keep credentials server-side, and operate browser, Android, recordings, schedules, integrations, memory, MCP, and messaging from one Flutter UI.
-  actions:
-    - theme: brand
-      text: Get Started
-      link: /getting-started
-    - theme: alt
-      text: Capabilities
-      link: /capabilities
-    - theme: alt
-      text: Why NeoAgent
-      link: /why-neoagent
+# NeoAgent
 
-features:
-  - title: Android Control
-    details: Let the AI operate a server-attached Android emulator or device with screenshots, UI dumps, app launch, intents, taps, typing, swipes, APK installs, and ADB shell.
-    link: /capabilities#android-control
-    linkText: Android Control
-  - title: Recordings
-    details: Capture web, Android, and wearable audio as sessions with transcripts, searchable segments, playback, retry, cleanup, and AI-generated insights.
-    link: /capabilities#recordings
-    linkText: Recordings
-  - title: Proactive Automation
-    details: Create recurring tasks and one-time runs that can use browser, files, CLI, memory, MCP, integrations, subagents, health summaries, and messaging delivery.
-    link: /automation
-    linkText: Automation
-  - title: Official Integrations
-    details: Use OAuth-backed Google Workspace, Microsoft 365, Notion, Slack, and Figma tools instead of brittle browser automation where possible.
-    link: /integrations
-    linkText: Integrations
-  - title: Server-Side Secrets
-    details: Keep AI provider keys, OAuth client secrets, Telnyx tokens, runtime settings, and deployment controls on the NeoAgent server.
-    link: /configuration
-    linkText: Configuration
-  - title: Recovery Path
-    details: Operate self-hosted installs with status, logs, release channels, update, fix, runtime paths, and the remote-server log caveat.
-    link: /operations
-    linkText: Operations
----
+NeoAgent is a self-hosted proactive AI agent with a bundled Flutter client for web and Android. It runs on your server, keeps credentials server-side, and gives you an operator UI for chat, runs, logs, scheduler tasks, skills, integrations, MCP, memory, Android control, recordings, Health Connect data, wearables, and settings.
+
+It is designed for people who want a focused personal automation server rather than a broad gateway platform. NeoAgent can run scheduled tasks, control a browser, operate a server-attached Android emulator or device, manage files, remember long-term context, connect to hosted AI providers or local Ollama, search recordings, read synced health summaries, and send results through Telegram, Discord, WhatsApp, or Telnyx Voice.
 
 ## Quick Start
 
@@ -52,7 +19,21 @@ neoagent install
 
 Open the server URL, sign in, configure providers and messaging, then create a scheduled task or chat run.
 
-## Navigation
+## What NeoAgent Does
+
+| Area | Capability |
+|---|---|
+| AI providers | OpenAI, Anthropic, xAI, Google, MiniMax Code, and local Ollama |
+| Operator UI | Chat, live runs, logs, scheduler, skills, integrations, MCP, memory, devices, recordings, health, wearables, settings |
+| Automation | Recurring scheduled tasks, one-time runs, browser control, file access, CLI skills, subagents, and messaging delivery |
+| Android control | AI control of a server-attached Android emulator or device: screenshots, UI dumps, taps, typing, intents, APK installs, and ADB shell |
+| Recordings | Web, Android, and wearable audio sessions with transcript search and AI insights |
+| Integrations | Google Workspace, Notion, Microsoft 365, Slack, Figma, and remote MCP servers |
+| Messaging | Telegram, Discord, WhatsApp text/media, and Telnyx Voice calls |
+| Outputs | Artifacts, Grok image generation, vision analysis, markdown tables, and Mermaid graphs |
+| Recovery | `neoagent status`, `neoagent logs`, `neoagent update`, release channels, and `neoagent fix` |
+
+## Main Paths
 
 | Need | Start here |
 |---|---|

package/package.json

@@ -1,9 +1,12 @@
 {
   "name": "neoagent",
-  "version": "2.1.18-beta.36",
+  "version": "2.1.18-beta.38",
   "description": "Proactive personal AI agent with no limits",
   "license": "MIT",
   "main": "server/index.js",
+  "engines": {
+    "node": ">=20"
+  },
   "bin": {
     "neoagent": "bin/neoagent.js"
   },
@@ -28,9 +31,9 @@
     "dev:stack": "./dev/stack.sh",
     "dev:build": "./dev/build.sh",
     "dev:test": "./dev/test.sh",
-    "docs:dev": "vitepress dev docs",
-    "docs:build": "vitepress build docs",
-    "docs:preview": "vitepress preview docs",
+    "docs:dev": "docusaurus start",
+    "docs:build": "docusaurus build",
+    "docs:preview": "docusaurus serve build",
     "flutter:run:web": "cd flutter_app && flutter run -d chrome",
     "flutter:build:web": "cd flutter_app && flutter build web --output ../server/public --dart-define=NEOAGENT_BACKEND_URL=${NEOAGENT_BACKEND_URL:-}",
     "manage": "node bin/neoagent.js",
@@ -83,6 +86,9 @@
     "undici": "^6.24.0"
   },
   "devDependencies": {
-    "vitepress": "1.6.4"
+    "@docusaurus/core": "3.8.1",
+    "@docusaurus/preset-classic": "3.8.1",
+    "react": "18.3.1",
+    "react-dom": "18.3.1"
   }
 }

package/server/public/flutter_bootstrap.js

@@ -37,6 +37,6 @@ _flutter.buildConfig = {"engineRevision":"425cfb54d01a9472b3e81d9e76fd63a4a44cfb
 
 _flutter.loader.load({
   serviceWorkerSettings: {
-    serviceWorkerVersion: "1390596255" /* Flutter's service worker is deprecated and will be removed in a future Flutter release. */
+    serviceWorkerVersion: "2247189506" /* Flutter's service worker is deprecated and will be removed in a future Flutter release. */
   }
 });

package/server/services/ai/engine.js

@@ -173,10 +173,10 @@ function joinSentMessages(messages = []) {
 
 function buildBlankMessagingReplyPrompt(attempt) {
   if (attempt <= 1) {
-    return 'You must send one non-empty plain-text reply for the external messaging user right now. Do not call tools. Do not use markdown. Give either: (a) the concrete outcome, or (b) a clear blocker and the next action. If tool work already happened, summarize what you actually tried and where it got blocked. Do not ask the user to repeat the original request.';
+    return 'You must send one non-empty plain-text reply for the external messaging user right now. Do not call tools. Do not use markdown. Give either: (a) the concrete outcome, or (b) a clear blocker. If tool work already happened, summarize what you actually tried and where it got blocked. Do not ask the user to repeat the original request. Do not promise future work unless that work already happened in this run or will happen automatically before this reply is sent.';
   }
 
-  return 'Your previous reply was empty. Return one non-empty plain-text message now. Do not call tools. Do not use markdown. If needed, apologize briefly, explain the blocker in one sentence, and tell the user what to do next. Use the run evidence already in the conversation instead of asking the user to restate the task.';
+  return 'Your previous reply was empty. Return one non-empty plain-text message now. Do not call tools. Do not use markdown. If needed, apologize briefly and explain the blocker in one sentence. Use the run evidence already in the conversation instead of asking the user to restate the task. Do not promise future work unless that work already happened in this run or will happen automatically before this reply is sent.';
 }
 
 function parseToolExecutionSummary(item) {
@@ -280,21 +280,35 @@ function buildDeterministicMessagingFallback({ failedStepCount, stepIndex, toolE
   return 'I could not produce a reliable final reply just now.';
 }
 
-function buildModelFailureLoopPrompt({ failedModel, nextModel, errorMessage }) {
-  return [
-    `The previous model call on "${failedModel}" failed with: ${summarizeForLog(errorMessage, 220)}.`,
-    `Continue on "${nextModel}" and recover autonomously.`,
-    'If a previous plan depended on that failed call, adjust your approach and proceed end-to-end.',
-    'Only ask the user for help if no safe path remains.'
-  ].join(' ');
-}
+function buildMessagingFailureScenario({ err, failedStepCount, stepIndex, toolExecutions = [] }) {
+  const parts = [];
+  const runtimeError = normalizeOutgoingMessage(err?.message || '');
+  const workSummary = summarizeRecentWork(toolExecutions);
+  const blocker = [...toolExecutions].reverse()
+    .map((item) => extractToolFailureMessage(item))
+    .find(Boolean);
 
-function buildMessagingErrorReply(err) {
-  const message = String(err?.message || '').trim();
-  if (!message) {
-    return 'I ran into an internal error while processing your request and could not finish it reliably.';
+  if (runtimeError) {
+    parts.push(`Runtime error: ${summarizeForLog(runtimeError, 260)}.`);
+  }
+  if (workSummary) {
+    parts.push(`Observed work before failure: ${workSummary}.`);
+  }
+  if (blocker) {
+    parts.push(`Most specific blocker from run evidence: ${summarizeForLog(blocker, 260)}.`);
+  }
+  if (stepIndex > 0) {
+    parts.push(`Completed steps before failure: ${stepIndex}.`);
+  }
+  if (failedStepCount > 0) {
+    parts.push(`Failed tool steps: ${failedStepCount}.`);
   }
 
+  return parts.join(' ');
+}
+
+function buildDeterministicMessagingErrorReply({ err, failedStepCount, stepIndex, toolExecutions = [] }) {
+  const message = normalizeOutgoingMessage(err?.message || '');
   if (/no ai providers? are currently available/i.test(message)) {
     return 'I cannot continue right now because no AI provider is available for this account. Please check the provider settings.';
   }
@@ -303,7 +317,27 @@ function buildMessagingErrorReply(err) {
     return 'I hit a timeout while processing your request and could not finish it reliably.';
   }
 
-  return 'I ran into an internal error while processing your request and could not finish it reliably.';
+  const blocker = [...toolExecutions].reverse()
+    .map((item) => extractToolFailureMessage(item))
+    .find(Boolean);
+  if (blocker) {
+    return `I got blocked while checking this: ${blocker}.`;
+  }
+
+  if (message) {
+    return `I got blocked while working on this: ${message}.`;
+  }
+
+  return buildDeterministicMessagingFallback({ failedStepCount, stepIndex, toolExecutions });
+}
+
+function buildModelFailureLoopPrompt({ failedModel, nextModel, errorMessage }) {
+  return [
+    `The previous model call on "${failedModel}" failed with: ${summarizeForLog(errorMessage, 220)}.`,
+    `Continue on "${nextModel}" and recover autonomously.`,
+    'If a previous plan depended on that failed call, adjust your approach and proceed end-to-end.',
+    'Only ask the user for help if no safe path remains.'
+  ].join(' ');
 }
 
 const MAX_AUTONOMOUS_MESSAGING_RETRIES = 2;
@@ -1984,12 +2018,18 @@ class AgentEngine {
         if (!runMeta?.messagingSent) {
           const manager = this.messagingManager;
           if (manager) {
+            const failureScenario = buildMessagingFailureScenario({
+              err,
+              failedStepCount,
+              stepIndex,
+              toolExecutions,
+            });
             try {
               const failedMessage = sanitizeConversationMessages([
                 ...messages,
                 {
                   role: 'system',
-                  content: `The run encountered a runtime error and cannot continue reliably: ${summarizeForLog(err.message, 260)}. Do not call tools. Write exactly one short plain-text user message that explains the blocker naturally. Do not ask the user to resend or restate the same task. Only ask the user for something if a specific external input, permission, or configuration change is actually required.`
+                  content: `The run encountered a runtime error and cannot continue reliably. Use the actual run scenario below to explain the blocker naturally.\n\nScenario:\n${failureScenario || 'No additional scenario details were captured.'}\n\nDo not call tools. Write exactly one short plain-text user message. Do not ask the user to resend or restate the same task. Only ask the user for something if a specific external input, permission, or configuration change is actually required. Do not promise future work unless it will happen automatically before this reply is sent.`
                 }
               ]);
               const modelReply = await provider.chat(failedMessage, [], {
@@ -2005,7 +2045,12 @@ class AgentEngine {
             }
 
             if (!messagingFailureContent) {
-              messagingFailureContent = buildMessagingErrorReply(err);
+              messagingFailureContent = buildDeterministicMessagingErrorReply({
+                err,
+                failedStepCount,
+                stepIndex,
+                toolExecutions,
+              });
             }
 
             try {

package/server/services/ai/systemPrompt.js

@@ -78,6 +78,7 @@ For long tasks, give brief progress only when the user is waiting or the operati
 REPORT ACTUAL RESULTS
 When a tool returns data, share the relevant parts — summarized if large, direct if short. Never paste raw JSON as the answer. Never narrate what you're about to do at length before doing it.
 Never promise an action in the final answer unless you already took that action in this run. Do not say "I'll check", "I'll fix it", or "I'll send it" and then stop. Either do it first or say you have not done it yet.
+Do not promise future follow-up work unless that work will actually happen automatically before the current run ends.
 For scheduler or task-config changes, never claim that a cron job was created, updated, deleted, enabled, disabled, or “fixed” unless the corresponding scheduler tool call succeeded in this run. If you did not verify the actual task config, say that clearly instead of guessing.
 If the user asks you to debug scheduler timing or frequency, inspect the current scheduled-task list first and separate three things clearly: what you observed, what you infer, and what you actually changed.
 

package/docs/.vitepress/config.mts

@@ -1,102 +0,0 @@
-import { defineConfig } from 'vitepress';
-
-export default defineConfig({
-  title: 'NeoAgent',
-  description: 'Self-hosted proactive AI agent docs',
-  base: '/NeoAgent/',
-  cleanUrls: true,
-  lastUpdated: true,
-  themeConfig: {
-    nav: [
-      {
-        text: 'Start',
-        activeMatch: '^/(getting-started|why-neoagent)?$',
-        items: [
-          { text: 'Overview', link: '/' },
-          { text: 'Getting Started', link: '/getting-started' },
-          { text: 'Why NeoAgent', link: '/why-neoagent' },
-        ],
-      },
-      {
-        text: 'Product',
-        activeMatch: '^/(capabilities|automation|integrations|skills)',
-        items: [
-          { text: 'Capabilities', link: '/capabilities' },
-          { text: 'Android Control', link: '/capabilities#android-control' },
-          { text: 'Recordings', link: '/capabilities#recordings' },
-          { text: 'Integrations', link: '/integrations' },
-          { text: 'Automation', link: '/automation' },
-        ],
-      },
-      {
-        text: 'Operate',
-        activeMatch: '^/(configuration|operations)',
-        items: [
-          { text: 'Configuration', link: '/configuration' },
-          { text: 'Skills', link: '/skills' },
-          { text: 'Operations', link: '/operations' },
-        ],
-      },
-      { text: 'Why NeoAgent', link: '/why-neoagent' },
-      { text: 'GitHub', link: 'https://github.com/NeoLabs-Systems/NeoAgent' },
-    ],
-    sidebar: [
-      {
-        text: 'Start',
-        items: [
-          { text: 'Overview', link: '/' },
-          { text: 'Getting Started', link: '/getting-started' },
-          { text: 'Why NeoAgent', link: '/why-neoagent' },
-        ],
-      },
-      {
-        text: 'Product Surface',
-        items: [
-          {
-            text: 'Capabilities',
-            link: '/capabilities',
-            items: [
-              { text: 'Android Control', link: '/capabilities#android-control' },
-              { text: 'Recordings', link: '/capabilities#recordings' },
-              { text: 'Health Data', link: '/capabilities#health-data' },
-              { text: 'Agent Tools', link: '/capabilities#agent-tools' },
-              { text: 'Runtime Modes', link: '/capabilities#runtime-modes' },
-            ],
-          },
-          { text: 'Automation', link: '/automation' },
-          { text: 'Integrations', link: '/integrations' },
-          { text: 'Skills', link: '/skills' },
-        ],
-      },
-      {
-        text: 'Operate',
-        items: [
-          { text: 'Configuration', link: '/configuration' },
-          { text: 'Operations', link: '/operations' },
-        ],
-      },
-    ],
-    socialLinks: [
-      { icon: 'github', link: 'https://github.com/NeoLabs-Systems/NeoAgent' },
-    ],
-    search: {
-      provider: 'local',
-    },
-    outline: {
-      level: [2, 3],
-      label: 'On This Page',
-    },
-    editLink: {
-      pattern: 'https://github.com/NeoLabs-Systems/NeoAgent/edit/main/docs/:path',
-      text: 'Edit this page on GitHub',
-    },
-    docFooter: {
-      prev: 'Previous',
-      next: 'Next',
-    },
-    footer: {
-      message: 'Released under the MIT License.',
-      copyright: 'Copyright NeoLabs Systems',
-    },
-  },
-});