admin/zCode-CLI-X
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: comprehensive README overhaul — streaming, self-correction, RTK, formatting

Browse Source 
- New streaming architecture section with ASCII pipeline diagrams
- Self-correction loops documented (2 retries + backoff + auto-simplification)
- RTK integration section (60-90% token savings)
- Telegram HTML formatting system documented
- Updated feature comparison table (all checkmarks accurate)
- systemd service deployment guide added
- Command reference table
- Architecture flow diagrams
This commit is contained in:
admin
2026-05-05 14:24:07 +00:00
Unverified
parent a0549ac641
commit 2a088cb0b9
1 changed files with 201 additions and 137 deletions
Download Patch File Download Diff File Expand all files Collapse all files

334
README.md
View File

@@ -1,19 +1,44 @@
# zCode CLI X # zCode CLI X
Agentic coder with **Z.AI + Telegram integration** — Claude Code + Hermes in one beast. Agentic coding assistant with **Z.AI + Telegram integration** — autonomous code execution with real-time streaming, self-correction loops, and RTK token optimization.
> 💡 **Get 10% OFF Z.AI** — Use code **ROK78RJKNW** at [z.ai/subscribe](https://z.ai/subscribe?ic=ROK78RJKNW) for the Coding Plan > 💡 **Get 10% OFF Z.AI** — Use code **ROK78RJKNW** at [z.ai/subscribe](https://z.ai/subscribe?ic=ROK78RJKNW) for the Coding Plan
## 🚀 Features ## ⚡ Features
### Core
- **🤖 AI-Powered Code Generation**: Powered by Z.AI GLM-5.1 (Coding Plan) - **🤖 AI-Powered Code Generation**: Powered by Z.AI GLM-5.1 (Coding Plan)
- **📱 Multi-Channel Bot**: Telegram + Discord with real-time delivery - **📱 Telegram Bot**: 24/7 via grammy + webhook with real-time SSE streaming
- **🛠️ Full Engineering Access**: Bash, FileEdit, WebSearch, Git tools - **🛠️ Full Engineering Access**: Bash, FileEdit, WebSearch, Git tools
- **🧠 Agent System**: Code reviewer, architect, DevOps engineer - **🧠 Agent System**: Code Reviewer, System Architect, DevOps Engineer
- **📚 Skills System**: Pre-built skills for common tasks - **📚 Skills System**: Pre-built skills for common tasks
- **⚡ Real-time Updates**: WebSocket-based live communication
- **🔄 Self-Correction**: Automatic retry with backoff (2 retries) ### Streaming & Formatting
- **📦 Multi-Channel Delivery**: Hub-based message routing (Telegram + Discord + WebSocket + log) - **⚡ Real-time SSE Streaming**: Token-by-token delivery via `StreamConsumer` — adapted from [Hermes Agent's GatewayStreamConsumer](https://github.com/nousresearch/hermes-agent)
- Queued token buffer → rate-limited `editMessageText` loop (1s base interval)
- Adaptive backoff on Telegram flood control (429)
- Typing cursor `▉` during generation, clean final message
- Graceful fallback to plain send on repeated failures
- **🎨 Telegram HTML Formatting**: AI markdown → clean Telegram HTML
- `**bold**`, `*italic*`, `` `code` ``, fenced code blocks, `[links](url)`, `~~strike~~`, headings, blockquotes, lists
- Double fallback: HTML → stripped plain text (never shows raw `**`)
### Reliability
- **🔄 Self-Correction Loops**: Automatic retry with exponential backoff
- 2 retry attempts (500ms → 1s → 1.5s delay)
- Triggers: API errors, rate limits, timeouts, 5xx server errors
- Auto-simplification: prompts simplified on retry to avoid recurring errors
- Full logging of all retry attempts with reason tracking
- **🛡️ RTK (Rust Token Killer)**: Token optimization for supported commands
- 60-90% savings on git, npm, cargo, pytest, docker, and more
- Active tracking stats via `getTrackingStats()`
### Architecture
- **📨 Multi-Channel Delivery**: Hub-based routing (Telegram + Discord + WebSocket + log)
- **🔁 Deduplication**: 60s TTL message deduplication
- **📋 Request Queue**: Per-chat sequential processing (no race conditions)
- **🔌 MCP Protocol**: Full MCP client + server management
- **⏰ Cron Scheduling**: 1s interval, task locking, auto-recovery
## 📦 Installation ## 📦 Installation
@@ -35,10 +60,10 @@ cp .env.example .env
```env ```env
# Z.AI Configuration (Coding Plan) # Z.AI Configuration (Coding Plan)
GLM_BASE_URL=https://api.z.ai/api/coding/paas/v4 GLM_BASE_URL=https://api.z.ai/api/coding/paas/v4
ZAI_API_KEY=your_z...here ZAI_API_KEY=***
# Telegram Bot Configuration # Telegram Bot Configuration
TELEGRAM_BOT_TOKEN=your_b...here TELEGRAM_BOT_TOKEN=***
TELEGRAM_ALLOWED_USERS=your_telegram_id TELEGRAM_ALLOWED_USERS=your_telegram_id
ZCODE_WEBHOOK_URL=https://your-domain.com/telegram/webhook ZCODE_WEBHOOK_URL=https://your-domain.com/telegram/webhook
``` ```
@@ -57,66 +82,115 @@ node bin/zcode.js
node bin/zcode.js --bot node bin/zcode.js --bot
``` ```
### Run in Development Mode ### Run as systemd service
```bash ```ini
node bin/zcode.js --dev # /etc/systemd/system/zcode.service
[Unit]
Description=zCode CLI X Bot
After=network.target
[Service]
Type=simple
User=uroma2
WorkingDirectory=/home/uroma2/zcode-cli-x
ExecStart=/usr/bin/node bin/zcode.js --bot
Restart=always
RestartSec=5
[Install]
WantedBy=multi-user.target
``` ```
### Run Only CLI (No Bot)
```bash ```bash
node bin/zcode.js --no-bot sudo systemctl enable zcode
sudo systemctl start zcode
``` ```
## 🤖 Telegram & Discord Bot ## 🤖 Telegram Bot Commands
Once running, you can interact with zCode CLI X via multiple channels: | Command | Description |
|---|---|
| `/start` | Show help and capabilities |
| `/tools` | List available tools |
| `/skills` | List loaded skills |
| `/agents` | List agent roles |
| `/model <name>` | Switch AI model |
| `/stats` | System & RTK stats |
| `/selfcorrection` | Self-correction status |
| `/bash <cmd>` | Execute shell command |
| `/web <query>` | Search the web |
| `/git <action>` | Git operations |
| `/cancel` | Cancel current operation |
### Telegram Or just chat — zCode uses tools automatically when needed.
1. Start the bot: `/start`
2. Send your code requests or questions
3. Receive AI-powered responses in real-time
### Discord ## 🛠️ Tools
1. Invite the bot to your server
2. Send commands or questions
3. Receive responses with full streaming support
### Multi-Channel Delivery | Tool | Description |
Messages are automatically delivered to: |---|---|
- ✅ Telegram (primary) | **BashTool** | Shell command execution with timeout control |
- ✅ Discord | **FileEditTool** | Diff-aware file operations (read, write, patch) |
- ✅ WebSocket (real-time) | **WebSearchTool** | Web search with result ranking |
- ✅ Log file (structured) | **GitTool** | Git operations (status, log, diff, commit, push, pull) |
## 🛠️ Tools Available
- **Bash**: Execute shell commands
- **FileEdit**: Edit files with diff-aware operations
- **WebSearch**: Search the web for information
- **Git**: Git operations (status, log, branch)
## 🧠 Agents ## 🧠 Agents
- **Code Reviewer**: Review code for bugs and improvements | Agent | Role |
- **System Architect**: Design system architecture and patterns |---|---|
- **DevOps Engineer**: Handle deployment, CI/CD, and infrastructure | **Code Reviewer** | Review code for bugs, security issues, and improvements |
| **System Architect** | Design system architecture and patterns |
## 📚 Skills | **DevOps Engineer** | Handle deployment, CI/CD, and infrastructure |
- **code_review**: Review code for bugs, security issues, and improvements
- **bug_fix**: Fix identified bugs in code
- **refactor**: Refactor code for better quality
- **documentation**: Generate and update documentation
- **testing**: Write tests for code
## 🏗️ Architecture ## 🏗️ Architecture
zCode CLI X uses a hybrid architecture combining: zCode CLI X uses a hybrid architecture:
- **better-clawd**: Coding backbone (BashTool, FileEditTool, WebSearchTool, GitTool)
- **Hermes Agent**: Agent system (Code Reviewer, Architect, DevOps) ```
- **claudegram**: Bot patterns (grammy, deduplication, request queuing, multi-channel delivery) User Message
│
▼
┌─────────────┐ ┌──────────────┐ ┌─────────────────┐
│ Telegram │────▶│ dedup.js │────▶│ request-queue │
│ (grammy) │ │ (60s TTL) │ │ (per-chat seq) │
└─────────────┘ └──────────────┘ └────────┬────────┘
│
▼
┌────────────────┐
│ self-correction │
│ (2 retries + │
│ backoff) │
└────────┬───────┘
│
┌────────────────────┼────────────────────┐
▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ chatWithAI │ │ Tool │ │ Agent │
│ (SSE stream) │ │ Handlers │ │ Delegation │
└──────┬───────┘ └──────────────┘ └──────────────┘
│
┌───────┴────────┐
▼ ▼
┌──────────────┐ ┌──────────────┐
│ Stream │ │ sendFormatted│
│ Consumer │ │ (HTML mode) │
│ (edit-in- │ └──────────────┘
│ place) │
└──────┬───────┘
│
▼
┌──────────────┐
│ markdown │
│ ToHtml() │
│ converter │
└──────┬───────┘
│
▼
┌──────────────┐
│ Telegram API │
│ (HTML mode) │
└──────────────┘
```
### Core Components ### Core Components
@@ -126,85 +200,71 @@ zcode-cli-x/
│ └── zcode.js # CLI entry point │ └── zcode.js # CLI entry point
├── src/ ├── src/
│ ├── bot/ │ ├── bot/
│ │ ├── index.js # Telegram bot (grammy-based) │ │ ├── index.js # Telegram bot (grammy + SSE streaming)
│ │ ├── message-sender.js # StreamConsumer + markdownToHtml converter
│ │ ├── deduplication.js # Message deduplication (60s TTL) │ │ ├── deduplication.js # Message deduplication (60s TTL)
│ │ ├── request-queue.js # Per-chat request queuing │ │ ├── request-queue.js # Per-chat request queuing
│ │ ├── message-sender.js # Message chunking (4k limit)
│ │ ├── delivery-hub.js # Multi-channel delivery │ │ ├── delivery-hub.js # Multi-channel delivery
│ │ ├── discord.js # Discord integration (discord.js v14) │ │ ├── discord.js # Discord integration (discord.js v14)
│ │ └── self-correction.js # Self-correction wrapper (2 retries) │ │ └── self-correction.js # Self-correction wrapper (2 retries + backoff)
│ ├── api/ │ ├── api/
│ │ └── index.js # Z.AI API adapter (GLM-5.1) │ │ └── index.js # Z.AI API adapter (GLM-5.1, SSE support)
│ ├── tools/ │ ├── tools/
│ │ ├── BashTool.js # Shell command executor │ │ ├── BashTool.js # Shell command executor (RTK-aware)
│ │ ├── FileEditTool.js # File operations │ │ ├── FileEditTool.js # File operations
│ │ ├── WebSearchTool.js # Web search │ │ ├── WebSearchTool.js # Web search
│ │ └── GitTool.js # Git operations │ │ └── GitTool.js # Git operations (RTK-aware)
│ ├── agents/ │ ├── agents/
│ │ └── index.js # Agent orchestration │ │ └── index.js # Agent orchestration
│ ├── skills/ │ ├── skills/
│ │ └── index.js # Skills system │ │ └── index.js # Skills system
│ └── utils/ │ └── utils/
│ ├── logger.js # Winston logger │ ├── logger.js # Winston logger
│ └── env.js # Environment validation │ ├── env.js # Environment validation
│ └── rtk.js # RTK (Rust Token Killer) integration
├── .env # Configuration ├── .env # Configuration
└── package.json └── package.json
``` ```
### Bot Flow ### Bot Message Flow
1. **Message Reception**: Telegram/Discord messages → `delivery-hub.js` 1. **Message Reception**: Telegram webhook → grammy handler
2. **Deduplication Check**: `deduplication.js` (60s TTL) 2. **Deduplication**: `deduplication.js` (60s TTL, prevents double-processing)
3. **Request Queue**: `request-queue.js` (per-chat sequential processing) 3. **Request Queue**: `request-queue.js` (per-chat sequential processing)
4. **Self-Correction**: `self-correction.js` (2 retries + backoff) 4. **Self-Correction**: `self-correction.js` (2 retries + exponential backoff + auto-simplification)
5. **Agent Execution**: better-clawd tools + Hermes agents 5. **AI Chat + Streaming**: `chatWithAI()` → SSE stream → `StreamConsumer` → real-time edits
6. **Response Delivery**: `message-sender.js` → Multi-channel (Telegram + Discord + WebSocket + log) 6. **Formatting**: `markdownToHtml()` converts AI markdown → Telegram HTML
7. **Final Delivery**: `editMessageText` with HTML parse_mode (or fallback to stripped plain text)
### Provider Support ### StreamConsumer Pipeline
- **Z.AI**: GLM-5.1 (Coding Plan) - Primary
- **Claude**: Anthropic Claude (via API)
- **OpenCode**: OpenCode CLI integration
- **Kilo**: Kilo.AI gateway
## 🔗 Integrations
- **Z.AI API**: GLM-5.1 model with Coding Plan
- **Telegram Bot API**: grammy-based bot with auto-retry and runner
- **Discord.js v14**: Discord bot with GatewayIntentBits
- **Express.js**: HTTP server for webhook handling
- **Winston**: Structured logging
- **WebSocket**: Real-time updates
- **Mongoose**: Database persistence
- **Puppeteer**: Browser automation
## 🚀 Getting Started
1. Install dependencies:
```bash
npm install
``` ```
Z.AI API (SSE)
2. Configure `.env` with your credentials: │
```env ▼ onDelta(token)
GLM_BASE_URL=https://api.z.ai/api/coding/paas/v4 ┌──────────────┐
ZAI_API_KEY=your_zai_api_key │ Token Buffer │ ← accumulates tokens from SSE stream
TELEGRAM_BOT_TOKEN=your_telegram_bot_token └──────┬───────┘
TELEGRAM_ALLOWED_USERS=your_telegram_user_id │ (every ~1s or 40 chars)
DISCORD_TOKEN=your_discord_bot_token ▼
┌──────────────┐
│ editMessage │ ← plain text + cursor ▉ (no parse_mode)
│ Text() │ rate-limited, adaptive backoff on flood
└──────┬───────┘
│ (on finish)
▼
┌──────────────┐
│ markdownTo │ ← converts **bold**, *italic*, `code`, etc.
│ Html() │ to <b>, <i>, <code>, <pre> HTML tags
└──────┬───────┘
│
▼
┌──────────────┐
│ editMessage │ ← final message with parse_mode: 'HTML'
│ Text() │ fallback: stripped plain text (no raw **)
└──────────────┘
``` ```
3. Run the bot:
```bash
node bin/zcode.js --bot
```
4. Open Telegram and Discord to start chatting!
## 📝 License
MIT
## 📊 Feature Comparison ## 📊 Feature Comparison
| Feature | zCode CLI X | Hermes Agent | better-clawd | | Feature | zCode CLI X | Hermes Agent | better-clawd |
@@ -213,47 +273,51 @@ MIT
| Autonomous execution | ✅ Full autonomous mode | ✅ Full autonomous mode | ⚠️ Manual step-by-step | | Autonomous execution | ✅ Full autonomous mode | ✅ Full autonomous mode | ⚠️ Manual step-by-step |
| Sub-agents | ✅ Multi-agent (swarm) | ✅ delegate_task + batch | ❌ Single agent only | | Sub-agents | ✅ Multi-agent (swarm) | ✅ delegate_task + batch | ❌ Single agent only |
| Agent roles | ✅ Code Reviewer, Architect, DevOps | ✅ Agent Registry (10+ roles) | ❌ Fixed single role | | Agent roles | ✅ Code Reviewer, Architect, DevOps | ✅ Agent Registry (10+ roles) | ❌ Fixed single role |
| Self-correction loops | ❌ None | ✅ Agent self-correction skill | ❌ None | | Self-correction loops | ✅ 2 retries + backoff + auto-simplification | ✅ Agent self-correction skill | ❌ None |
| **Streaming** | | | |
| Real-time SSE streaming | ✅ StreamConsumer (edit-in-place) | ✅ GatewayStreamConsumer | ❌ None |
| Telegram HTML formatting | ✅ markdownToHtml + fallback | ✅ Native HTML support | ❌ None |
| Adaptive flood control | ✅ Exponential backoff | ✅ Flood backoff | ❌ N/A |
| **Tooling** | | | | | **Tooling** | | | |
| Bash/Shell | ✅ BashTool | ✅ TerminalTool | ✅ Shell access | | Bash/Shell | ✅ BashTool | ✅ TerminalTool | ✅ Shell access |
| File editing | ✅ FileEditTool (diff-aware) | ✅ Patch + Write + Edit | ⚠️ Basic write | | File editing | ✅ FileEditTool (diff-aware) | ✅ Patch + Write + Edit | ⚠️ Basic write |
| Web search | ✅ WebSearch | ✅ WebSearch + Vane + Exa | ❌ None | | Web search | ✅ WebSearch | ✅ WebSearch + Vane + Exa | ❌ None |
| Git integration | ✅ GitTool | ✅ GitTool | ❌ None | | Git integration | ✅ GitTool (RTK-aware) | ✅ GitTool | ❌ None |
| Browser automation | ✅ Computer-use (Anthropic) | ✅ Full browser toolkit | ❌ None | | Browser automation | ✅ Computer-use (Anthropic) | ✅ Full browser toolkit | ❌ None |
| MCP servers | ✅ Full MCP protocol (client + server management) | ✅ Native MCP + mcporter | ❌ None | | MCP servers | ✅ Full MCP protocol | ✅ Native MCP + mcporter | ❌ None |
| Code execution | ✅ Sandbox adapter | ✅ Sandbox + Jupyter | ❌ None | | RTK optimization | ✅ RTK active (60-90% savings) | ✅ RTK integrated | ❌ None |
| **Skills** | | | |
| Skill system | ✅ Skill system with skills dir loader | ✅ 500+ skills catalog | ❌ No skill system |
| Custom skill authoring | ✅ Skillify (session→skill capture) | ✅ skill_manage CLI | ❌ None |
| Plugin architecture | ✅ Full marketplace + loader + lifecycle | ✅ Full plugin system | ❌ None |
| **Automation** | | | |
| Cron scheduling | ✅ CronScheduler (1s interval, jitter, locks) | ✅ Cron jobs with delivery | ❌ None |
| Webhook subscriptions | ✅ Hook system (HTTP, agent, prompt hooks) | ✅ Event-driven agent runs | ❌ None |
| Scheduled monitoring | ✅ Cron-based recurring monitoring | ✅ Browser + social monitors | ❌ None |
| Batch task processing | ✅ Batch skill (5-30 parallel subagents) | ✅ Parallel batch delegation | ❌ None |
| **Platform** | | | | | **Platform** | | | |
| Telegram integration | ✅ Native bot + webhook | ✅ 2-way Telegram bridge | ❌ None | | Telegram integration | ✅ Native bot + webhook + streaming | ✅ 2-way Telegram bridge | ❌ None |
| Discord | ✅ Native bot (via discord.js) | ✅ Full Discord integration | ❌ None | | Discord | ✅ Native bot (discord.js) | ✅ Full Discord integration | ❌ None |
| Multi-channel delivery | ✅ Delivery hub (Telegram + Discord + WebSocket + log) | ✅ Cron→Telegram/Discord/Email | ❌ None | | Multi-channel delivery | ✅ Delivery hub (TG + DC + WS + log) | ✅ Cron→multi-platform | ❌ None |
| Self-correction loops | ✅ Self-correction wrapper (2 retries + backoff) | ✅ Agent self-correction skill | ❌ None |
| **Infrastructure** | | | | | **Infrastructure** | | | |
| Model routing | ✅ Multi-provider (OpenAI, Anthropic, Bedrock, custom) | ✅ Multi-provider routing | ❌ Single model | | Model routing | ✅ Multi-provider | ✅ Multi-provider routing | ❌ Single model |
| Context compression | ✅ Compact pipeline (auto, micro, session memory) | ✅ lean-ctx MCP (90% savings) | ❌ None | | Context compression | ✅ Compact pipeline | ✅ lean-ctx MCP (90% savings) | ❌ None |
| Memory persistence | ✅ Session memory with background extraction | ✅ Cross-session memory | ❌ None | | Memory persistence | ✅ Session memory | ✅ Cross-session memory | ❌ None |
| RTK integration | ✅ RTK active | ✅ RTK integrated | ❌ None | | Cron scheduling | ✅ 1s interval, jitter, locks | ✅ Cron jobs with delivery | ❌ None |
### Summary ### Summary
- **zCode CLI X** — Lightweight agentic coder focused on Telegram + Z.AI. Ideal for quick coding tasks via Telegram with essential tools and agent roles. - **zCode CLI X** — Lightweight agentic coder focused on Telegram + Z.AI. Real-time SSE streaming, self-correction loops, RTK optimization, and beautiful HTML formatting. Ideal for quick coding tasks via Telegram.
- **Hermes Agent** — Full-stack AI assistant platform. Best for complex multi-agent workflows, scheduled automation, and cross-platform deployment. 500+ skills, MCP ecosystem, and the deepest toolset. - **Hermes Agent** — Full-stack AI assistant platform. Best for complex multi-agent workflows, scheduled automation, and cross-platform deployment. 500+ skills, MCP ecosystem, deepest toolset.
- **better-clawd** — Minimal Claude Code clone. Useful as a lightweight reference but lacks the agentic, skills, and automation depth of zCode or Hermes. - **better-clawd** — Minimal Claude Code clone. Useful as a lightweight reference but lacks agentic depth.
## 🔗 Integrations
- **Z.AI API**: GLM-5.1 model (Coding Plan) with SSE streaming
- **Telegram Bot API**: grammy + auto-retry + runner + webhook
- **Discord.js v14**: Discord bot with GatewayIntentBits
- **Express.js**: HTTP server for webhook handling
- **Winston**: Structured logging
- **WebSocket**: Real-time updates
- **RTK**: Rust Token Killer (token optimization)
## 🤝 Contributing ## 🤝 Contributing
Contributions welcome! This is based on: Contributions welcome! Based on:
- [better-clawd](https://github.com/x1xhlol/better-clawd.git) - Claude Code clone - [better-clawd](https://github.com/x1xhlol/better-clawd.git) — Claude Code clone
- [Hermes Agent](https://hermes-agent.nousresearch.com) - Our AI assistant - [Hermes Agent](https://hermes-agent.nousresearch.com) — AI assistant platform (streaming architecture credit)
--- ---
Built with ❤️ by zCode CLI X Built with ⚡ by zCode CLI X