zyn-ai 1.3.6 → 1.4.0

package/README.md

@@ -8,42 +8,41 @@
   <img src="https://img.shields.io/npm/v/zyn-ai?label=npm&color=%23CB3837" alt="NPM Version"/>
   <img src="https://img.shields.io/github/v/release/SoyMaycol/Zyn?include_prereleases&sort=semver" alt="Latest Release"/>
   <img src="https://img.shields.io/npm/dt/zyn-ai" alt="Downloads"/>
+  <img src="https://img.shields.io/github/license/SoyMaycol/Zyn" alt="License"/>
 </p>
 
 <p align="center">
-  <b>Local AI agent for terminal, TUI, and web.</b>
+  <b>A complete CLI Agent for production and enterprise level, with incredible results.</b>
 </p>
 
 <p align="center">
-  <a href="https://github.com/SoyMaycol/Zyn">Official repository</a>
+  <a href="https://github.com/SoyMaycol/Zyn">GitHub</a>
 </p>
 
 ---
 
-## What is Zyn
+## Features
 
-Zyn is a local AI agent designed for terminal and web usage. It supports persistent sessions, system tools, multiple AI providers, session exports, and configurable models.
-
----
-
-## Requirements
-
-- Node.js 18+
-- npm
-- Internet connection for remote providers
+- **CLI + TUI** — full terminal UI with keyboard navigation, token display, **improved history & compaction**, overlay system
+- **Multi-provider** — Zen (free, no config), Gemini, Qwen (DashScope), HuggingFace, **skill search**, custom providers (any OpenAI-compatible API)
+- **Skills system** — folder-based skills with **improved prompt integration** and YAML frontmatter
+- **Tool execution** — read/write files, run commands, search code, browse web, glob patterns
+- **Session management** — **real conversation history**, persistent sessions with **better compaction**, full transcript replay, resume, export
+- **Context tracking** — token estimator with per-model context limits, **auto-compaction at 85%**, updated to reflect new history behavior
+- **Multi-platform** — embeddable in WhatsApp (Baileys), Discord, and Telegram bots, Although this function is not complete (It is under development)
+- **Background workers** — detach long-running turns to background processes
+- **i18n** — English and Spanish interfaces
 
 ---
 
 ## Installation
 
-### Global install
-
 ```bash
 npm install -g zyn-ai
 zyn
 ```
 
-### Local development
+### From source
 
 ```bash
 git clone https://github.com/SoyMaycol/Zyn.git
@@ -52,125 +51,190 @@ npm install
 npm start
 ```
 
----
-
 ## Usage
 
 ```bash
-zyn
-zyn "Explain this project"
-zyn --new
-zyn --resume ID
+zyn                  # Interactive TUI (default)
+zyn "question"       # Single prompt (CLI mode)
+zyn --new            # Force new session
+zyn --resume ID      # Resume existing session (keeps history)
 ```
 
----
-
-## Web mode
-
-Inside Zyn:
+## Models
 
-```text
-/web
-/web 0.0.0.0:3000
-```
+14 built-in models across 4 providers. Each model has its real context length in the StatusBar (e.g. `10.8K/128K`).
 
-Or directly:
+### Zen (free, no configuration)
 
-```bash
-npm run web
-```
+| Key | Model | Context |
+|---|---|---|
+| `nemotron` | Nemotron 3 Ultra | 128K |
+| `mimo` | Mimo 2.5 | 128K |
+| `north-mini` | North Mini Code | 128K |
+| `deepseek` | DeepSeek V4 Flash | 128K |
 
----
+### Gemini (requires API key)
 
-## Language
+| Key | Model | Context |
+|---|---|---|
+| `gemini-flash` | Gemini 2.5 Flash | 1M |
+| `gemini-flash-001` | Gemini 2.5 Flash 001 | 1M |
+| `gemini-pro` | Gemini 2.5 Pro | 1M |
+| `gemini-flash-lite` | Gemini 2.5 Flash Lite | 1M |
+| `gemini-flash-lite-001` | Gemini 2.5 Flash Lite 001 | 1M |
+| `gemma-3` | Gemma 3 27B | 128K |
 
-Supported languages:
+### Qwen (requires DashScope API key)
 
-- `en`
-- `es`
+| Key | Model | Context |
+|---|---|---|
+| `qwen-plus` | Qwen Plus | 128K |
+| `qwen-max` | Qwen Max | 32K |
+| `qwen-turbo` | Qwen Turbo | 1M |
 
-Commands:
+### HuggingFace (requires HF token)
 
-```text
-/lang
-/lang en
-/lang es
-```
+| Key | Model | Context |
+|---|---|---|
+| `hf-ling-2.6-1t` | InclusionAI Ling 2.6 1T | 128K |
 
----
+Default model: `nemotron` (Zen, no configuration required).
 
-## Main Commands
+## Commands
 
 ### Sessions
 
 | Command | Description |
 |---|---|
-| `/help` | Show available commands |
-| `/status` | Show current status |
-| `/history` | Show recent actions |
-| `/memory` | Show memory summary |
-| `/sessions` | List saved sessions |
-| `/new` | Create a new session |
-| `/resume <ID>` | Resume a session |
+| `/help` | Show help |
+| `/status` | Current status |
+| `/history` | Recent actions (last 20) |
+| `/memory` | Memory summary |
+| `/session` | Current session info |
+| `/sessions` | List all sessions |
+| `/new` | New session |
+| `/resume <ID>` | Resume session (keeps history) |
 | `/title <text>` | Rename session |
 
 ### Configuration
 
 | Command | Description |
 |---|---|
-| `/model` | Show or change model |
-| `/models` | List models |
-| `/providers` | List providers |
+| `/models` | Open model picker (current provider) |
+| `/providers` | Open interactive provider picker → configure → pick model |
+| `/provider list` | List configured providers and their fields |
+| `/provider sync <name>` | Fetch models from a provider's API |
+| `/provider set <name> <field> <value>` | Set provider config (apiKey, baseUrl, modelId, contextLength) |
+| `/provider remove <name>` | Remove a provider configuration |
 | `/lang <en\|es>` | Change language |
+| `/auto on\|off` | Toggle auto-approval |
+| `/concuerdo` | Toggle group model mode (queries all configured models) |
+| `/persona set <text>` | Set response persona |
 | `/config show` | Show config |
-| `/auto on\|off` | Toggle auto approval |
+| `/git set\|list\|remove` | Manage git credentials |
 | `/cwd <path>` | Change working directory |
 
-### Tools
+### Tools & Skills
 
 | Command | Description |
 |---|---|
-| `/tools` | List tools |
-| `/skills` | List skills |
-| `/gmail connect` | Connect Gmail with Google OAuth + PKCE |
-| `/gmail status` | Show Gmail connection status |
-| `/gmail disconnect` | Remove saved Gmail tokens |
-| `/cwd` | Show working directory |
+| `/tools` | List agent tools |
+| `/skills` | List loaded skills |
+| `/gmail connect` | Connect Gmail via OAuth |
 
-### Web & Export
+### Export & Control
 
 | Command | Description |
 |---|---|
-| `/web` | Start web interface |
-| `/transcript` | Show transcript |
-| `/export` | Export session |
+| `/bg` | Detach current turn to background |
+| `/transcript` | View session transcript |
+| `/export` | Export session to txt |
+| `/stop` | Stop current agent turn |
+| `/undo` | Undo last turn |
+| `/redo` | Redo undone turn |
+| `/reset` | Reset context |
+| `/exit` | Exit |
 
-### Control
+Press `ESC` twice in the TUI to stop the current task.
 
-| Command | Description |
-|---|---|
-| `/stop` | Stop current task |
-| `/reset` | Reset session |
-| `/exit` | Exit Zyn |
+## Custom Providers
 
-In the TUI, press `ESC` twice to stop the current task.
+Add any OpenAI-compatible API:
 
+```bash
+# Interactive: run /providers → select "+ Add custom provider"
+# Or manual:
+/provider set groq baseUrl https://api.groq.com/openai/v1
+/provider set groq apiKey gsk_xxxx
+/provider set groq contextLength 128000
+/provider sync groq
+```
+
+Configurable fields: `apiKey`, `baseUrl`, `modelId`, `contextLength`, `email`, `password`, `modelEndpoint`, `chatEndpoint`.
+
+## Skills
+
+Skills are folders under `data/skills/<name>/SKILL.md` with YAML frontmatter:
+
+```markdown
+---
+name: reasoning
+description: Reasoning and planning for complex tasks
 ---
 
-## Models
+# Skill body here
+```
+
+The system prompt automatically advertises every loaded skill to the model, and mentions of GitHub repos/skills trigger `web_search` to find them.
+
+## API
 
-Custom models can be added using `data/models.json`.
+### Embed in your bot
 
-Example:
+```js
+const { createAgent } = require('zyn-ai');
 
-```json
-{
-  "models": {
-    "my-gemini-flash": {
-      "label": "Gemini Flash",
-      "provider": "gemini",
-      "geminiModel": "gemini-flash"
-    }
-  }
-}
+const agent = createAgent({
+  model: 'nemotron',
+  language: 'en',
+  autoApprove: false,
+});
+
+const response = await agent.send('userId', 'Hello!');
 ```
+
+### Platforms
+
+```js
+const { createAgent, platforms } = require('zyn-ai');
+
+const agent = createAgent({ model: 'nemotron' });
+
+// WhatsApp (Baileys)
+await platforms.whatsapp({ agent, session: './whatsapp-auth' });
+
+// Discord
+await platforms.discord({ agent, token: 'DISCORD_TOKEN' });
+
+// Telegram
+await platforms.telegram({ agent, token: 'TELEGRAM_TOKEN' });
+```
+
+Install the corresponding `optionalDependencies` only if you need them.
+
+## Environment Variables
+
+| Variable | Description |
+|---|---|
+| `ZYN_DEFAULT_MODEL` | Override default model key |
+| `ZYN_DEFAULT_LANG` | Default language (`en` or `es`) |
+| `ZYN_GEMINI_API_KEY` | Gemini API key |
+| `ZYN_QWEN_API_KEY` | DashScope API key |
+| `ZYN_HUGGINGFACE_TOKEN` | HuggingFace token |
+| `ZYN_REQUEST_TIMEOUT_MS` | Request timeout (default: 180000) |
+| `ZYN_PROVIDER_TIMEOUT_MAX_ATTEMPTS` | Retry attempts on provider failure (default: 3) |
+| `ZYN_GMAIL_CLIENT_SECRET` | Gmail OAuth client secret |
+
+## License
+
+MIT - Maycol B.T

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "zyn-ai",
-  "version": "1.3.6",
-  "description": "Production-ready AI agent for CLI and web with real tool execution and automation",
+  "version": "1.4.0",
+  "description": "Production-ready AI agent for CLI, TUI, and external platforms (WhatsApp/Baileys, Discord, Telegram) with real tool execution and automation",
   "author": "Maycol",
   "keywords": [
     "ai-agent",
@@ -14,7 +14,12 @@
     "productivity-tool",
     "local-ai-agent",
     "command-line-ai",
-    "web-ai-agent",
+    "tui-agent",
+    "whatsapp-bot",
+    "discord-bot",
+    "telegram-bot",
+    "multi-platform",
+    "embeddable-agent",
     "zyn"
   ],
   "bin": {
@@ -22,23 +27,23 @@
     "Zyn": "./zyn.js"
   },
   "type": "commonjs",
+  "main": "./src/agent.js",
   "scripts": {
     "start": "node zyn.js",
     "dev": "node zyn.js",
-    "web": "node src/web/server.js",
-    "check": "node --check zyn.js && node --check src/cli/runtime.js"
+    "check": "node --check zyn.js && node --check src/cli/runtime.js && node --check src/agent.js"
   },
   "dependencies": {
     "axios": "^1.7.9",
-    "bcryptjs": "^3.0.3",
     "cheerio": "^1.2.0",
-    "express": "^5.2.1",
-    "express-session": "^1.19.0",
     "ink": "^6.8.0",
-    "keypress": "^0.2.1",
     "react": "^19.0.0",
-    "session-file-store": "^1.5.0",
-    "jimp": "^1.6.1"
+    "jimp": "^0.16.1"
+  },
+  "optionalDependencies": {
+    "@whiskeysockets/baileys": "^6.7.0",
+    "discord.js": "^14.16.0",
+    "node-telegram-bot-api": "^0.66.0"
   },
   "repository": {
     "type": "git",
@@ -59,7 +64,12 @@
     "LICENSE"
   ],
   "exports": {
-    ".": "./zyn.js"
+    ".": "./src/agent.js",
+    "./agent": "./src/agent.js",
+    "./platforms": "./src/platforms/index.js",
+    "./platforms/baileys": "./src/platforms/baileys.js",
+    "./platforms/discord": "./src/platforms/discord.js",
+    "./platforms/telegram": "./src/platforms/telegram.js"
   },
   "preferGlobal": true
 }

package/src/agent.js

@@ -0,0 +1,168 @@
+const { EventEmitter } = require('events');
+const { runAgentTurn } = require('./core/agent');
+const { createNewSessionState, loadSessionState, loadOrCreateSessionState, listSessions, saveState } = require('./utils/sessionStorage');
+const { MODELS, DEFAULT_MODEL_KEY, DEFAULT_LANGUAGE } = require('./config');
+const { listSkills, listModels } = require('./public/helpers');
+const { runBackgroundWorker, detachBackgroundTurn } = require('./utils/backgroundWorker');
+const { enqueueBackgroundTask, listBackgroundResults, consumeBackgroundResult } = require('./utils/sessionStorage');
+
+function noopUi() {
+  return {
+    beginThinkingStream() {},
+    writeThinkingDelta() {},
+    endThinkingStream() {},
+    beginAssistantStream() {},
+    writeAssistantDelta() {},
+    endAssistantStream() {},
+    startThinkingIndicator() { return () => {}; },
+    pushAction() {},
+    logEvent() {},
+    paint: (text) => text,
+  };
+}
+
+function createAgent(options = {}) {
+  const emitter = new EventEmitter();
+  emitter.setMaxListeners(50);
+
+  const config = {
+    sessionId: options.sessionId || null,
+    cwd: options.cwd || process.cwd(),
+    model: options.model || DEFAULT_MODEL_KEY,
+    language: options.language || DEFAULT_LANGUAGE,
+    autoApprove: options.autoApprove !== undefined ? Boolean(options.autoApprove) : true,
+    personaPrompt: options.personaPrompt || '',
+    userId: options.userId || 'default',
+    resume: options.resume || false,
+  };
+
+  let state = null;
+  let initPromise = null;
+
+  async function ensureState() {
+    if (state) return state;
+    if (!initPromise) {
+      initPromise = (async () => {
+        if (config.sessionId) {
+          const loaded = await loadSessionState(config.sessionId, null);
+          if (loaded) {
+            if (config.cwd) loaded.cwd = config.cwd;
+            loaded.activeModel = config.model;
+            loaded.language = config.language;
+            loaded.autoApprove = config.autoApprove;
+            if (config.personaPrompt) loaded.personaPrompt = config.personaPrompt;
+            state = loaded;
+            return state;
+          }
+        }
+        if (config.resume) {
+          const loaded = await loadOrCreateSessionState(null, { forceNew: false, resume: true });
+          state = loaded.state;
+          return state;
+        }
+        state = await createNewSessionState(null);
+        state.cwd = config.cwd;
+        state.activeModel = config.model;
+        state.language = config.language;
+        state.autoApprove = config.autoApprove;
+        state.personaPrompt = config.personaPrompt;
+        await saveState(state);
+        return state;
+      })();
+    }
+    return initPromise;
+  }
+
+  emitter.send = async function send(userId, text) {
+    const s = await ensureState();
+    const userKey = userId || config.userId;
+    s.history.push({ role: 'user', content: String(text), userId: userKey });
+    const ui = noopUi();
+    const controller = new AbortController();
+    emitter.emit('turnStart', { userId: userKey, text });
+    try {
+      const result = await runAgentTurn(String(text), s, ui, { signal: controller.signal });
+      emitter.emit('turnEnd', { userId: userKey, text, content: result.content });
+      emitter.emit('message', { userId: userKey, role: 'assistant', content: result.content });
+      return result.content;
+    } catch (err) {
+      emitter.emit('error', err);
+      throw err;
+    } finally {
+      await saveState(s);
+    }
+  };
+
+  emitter.detach = function detach(userId, text) {
+    return (async () => {
+      const s = await ensureState();
+      const userKey = userId || config.userId;
+      const taskId = enqueueBackgroundTask({ sessionId: s.sessionId, input: String(text), detachedAt: new Date().toISOString() });
+      detachBackgroundTurn({
+        taskId,
+        sessionId: s.sessionId,
+        input: String(text),
+        cwd: s.cwd,
+        modelKey: s.activeModel,
+        language: s.language,
+        personaPrompt: s.personaPrompt,
+        autoApprove: s.autoApprove,
+      });
+      return taskId;
+    })();
+  };
+
+  emitter.consumeBackground = async function consumeBackground() {
+    const s = await ensureState();
+    const results = await listBackgroundResults(s.sessionId);
+    for (const r of results) {
+      emitter.emit('background', r);
+      await consumeBackgroundResult(r.taskId);
+    }
+    return results;
+  };
+
+  emitter.getState = async function getState() {
+    const s = await ensureState();
+    return {
+      sessionId: s.sessionId,
+      title: s.title,
+      cwd: s.cwd,
+      activeModel: s.activeModel,
+      language: s.language,
+      turnCount: s.turnCount,
+      historyLength: s.history.length,
+    };
+  };
+
+  emitter.reset = async function reset() {
+    const s = await ensureState();
+    s.history = [];
+    s.actionLog = [];
+    s.turnCount = 0;
+    s.memorySummary = '';
+    await saveState(s);
+    return true;
+  };
+
+  emitter.close = async function close() {
+    if (state) {
+      await saveState(state);
+    }
+  };
+
+  return emitter;
+}
+
+module.exports = {
+  createAgent,
+  listSkills,
+  listModels,
+  listSessions,
+  loadSessionState,
+  runBackgroundWorker,
+  runAgentTurn,
+  DEFAULT_LANGUAGE,
+  DEFAULT_MODEL_KEY,
+  MODELS,
+};