@shareai-lab/kode 1.0.69
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/LICENSE +661 -0
- package/README.md +113 -0
- package/cli.mjs +1803 -0
- package/package.json +92 -0
- package/yoga.wasm +0 -0
package/README.md
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
1
|
+
# Kode With Agent
|
|
2
|
+
|
|
3
|
+
You can see this as open-cc for everyone, agent-system design is in [./system-design.md](./system-design.md)
|
|
4
|
+
|
|
5
|
+
## Features
|
|
6
|
+
|
|
7
|
+
- ๐ ๏ธ **Code Analysis & Fixes** - Analyzes and improves your codebase
|
|
8
|
+
- ๐ **Code Explanation** - Explains complex functions and logic
|
|
9
|
+
- ๐งช **Test Execution** - Runs tests and shell commands
|
|
10
|
+
- ๐ง **Workflow Automation** - Handles entire development workflows
|
|
11
|
+
- ๐ค **Multi-Model Support** - Works with any OpenAI-compatible API
|
|
12
|
+
- ๐ฏ **Many Built-in Tools** - File operations, shell execution, notebooks, and more
|
|
13
|
+
- ๐พ **Smart Checkpoints** - Intelligent project state management and recovery
|
|
14
|
+
- ๐ฟ **Worktree Workflows** - Isolated development environments for features
|
|
15
|
+
|
|
16
|
+
## Installation
|
|
17
|
+
|
|
18
|
+
```bash
|
|
19
|
+
# not relased, i am on the plane with very poor network, if the plane can not land, anyone see this help me make this agent-system more strong.
|
|
20
|
+
```
|
|
21
|
+
|
|
22
|
+
## Quick Start
|
|
23
|
+
|
|
24
|
+
1. **Model Setup**: Use the onboarding flow or `/model` command to configure your AI provider
|
|
25
|
+
2. **Custom Models**: If your model isn't listed, manually configure it via `/config`
|
|
26
|
+
3. **OpenAI-Compatible**: Works with any OpenAI-style endpoint (Ollama, OpenRouter, etc.)
|
|
27
|
+
|
|
28
|
+
## MCP Server Integration
|
|
29
|
+
|
|
30
|
+
Use Agent Kode as a Model Context Protocol server with Claude Desktop:
|
|
31
|
+
|
|
32
|
+
1. Find the full path: `which kode`
|
|
33
|
+
2. Add to Claude Desktop config:
|
|
34
|
+
```json
|
|
35
|
+
{
|
|
36
|
+
"mcpServers": {
|
|
37
|
+
"agent-kode": {
|
|
38
|
+
"command": "/path/to/kode",
|
|
39
|
+
"args": ["mcp", "serve"]
|
|
40
|
+
}
|
|
41
|
+
}
|
|
42
|
+
}
|
|
43
|
+
```
|
|
44
|
+
|
|
45
|
+
## Development
|
|
46
|
+
|
|
47
|
+
```bash
|
|
48
|
+
# Install dependencies
|
|
49
|
+
pnpm install
|
|
50
|
+
|
|
51
|
+
# Run in development mode
|
|
52
|
+
pnpm run dev
|
|
53
|
+
|
|
54
|
+
# Build for production
|
|
55
|
+
pnpm run build
|
|
56
|
+
|
|
57
|
+
# Debug with verbose logging
|
|
58
|
+
NODE_ENV=development pnpm run dev --verbose --debug
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
## Architecture
|
|
62
|
+
|
|
63
|
+
- **React/Ink** - Terminal UI framework
|
|
64
|
+
- **18 Core Tools** - File operations, shell execution, AI workflows, checkpoints
|
|
65
|
+
- **Multi-Provider** - Anthropic Claude, OpenAI, custom endpoints
|
|
66
|
+
- **TypeScript** - Full type safety throughout
|
|
67
|
+
- **MCP Compatible** - Model Context Protocol integration
|
|
68
|
+
- **Smart Workflows** - Checkpoint system and worktree management
|
|
69
|
+
|
|
70
|
+
## Advanced Workflows
|
|
71
|
+
|
|
72
|
+
Agent Kode provides sophisticated development workflow management:
|
|
73
|
+
|
|
74
|
+
### ๐ฏ Checkpoint System
|
|
75
|
+
Intelligent project state management with automatic analysis and recovery:
|
|
76
|
+
- **`/checkpoint-save`** - Smart analysis and state preservation
|
|
77
|
+
- **`/checkpoint-restore`** - Natural language version recovery
|
|
78
|
+
|
|
79
|
+
### ๐ฟ Worktree Development
|
|
80
|
+
Isolated development environments for feature work:
|
|
81
|
+
- **`/worktree-create`** - Task-driven environment creation
|
|
82
|
+
- **`/worktree-review`** - Comprehensive code quality assessment
|
|
83
|
+
- **`/worktree-merge`** - Safe integration with quality gates
|
|
84
|
+
|
|
85
|
+
๐ **[Complete Workflow Documentation](docs/commands/README.md)**
|
|
86
|
+
|
|
87
|
+
## Bug Reports
|
|
88
|
+
|
|
89
|
+
Submit bugs directly from the app using `/bug` - it will open GitHub with pre-filled information.
|
|
90
|
+
|
|
91
|
+
## Privacy & Data
|
|
92
|
+
|
|
93
|
+
- **No telemetry** - No backend servers except your chosen AI providers
|
|
94
|
+
- **Local processing** - All data stays on your machine
|
|
95
|
+
- **Open source** - Full transparency in code and data handling
|
|
96
|
+
|
|
97
|
+
## Repository
|
|
98
|
+
|
|
99
|
+
- **Homepage**: [https://github.com/shareAI-lab/agent-kode](https://github.com/shareAI-lab/agent-kode)
|
|
100
|
+
- **Issues**: [https://github.com/shareAI-lab/agent-kode/issues](https://github.com/shareAI-lab/agent-kode/issues)
|
|
101
|
+
|
|
102
|
+
## License
|
|
103
|
+
|
|
104
|
+
See [LICENSE.md](LICENSE.md) for details.
|
|
105
|
+
|
|
106
|
+
---
|
|
107
|
+
|
|
108
|
+
**โ ๏ธ Use at your own risk** - This tool executes code and commands on your system.
|
|
109
|
+
|
|
110
|
+
## Thanks
|
|
111
|
+
- some code from @dnakov 's anonkode
|
|
112
|
+
- some ui learn from gemini-cli
|
|
113
|
+
- some system design learn from claude code
|