ltcai 0.1.22 → 0.1.24
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/README.md +227 -256
- package/docs/CHANGELOG.md +79 -1
- package/docs/images/logo.svg +33 -0
- package/docs/images/screenshot-admin.png +0 -0
- package/docs/images/screenshot-chat.png +0 -0
- package/docs/images/screenshot-graph.png +0 -0
- package/package.json +1 -2
- package/server.py +17 -4
- package/tests/__init__.py +0 -0
- package/tests/integration/__init__.py +0 -0
- package/tests/integration/test_api.py +0 -94
- package/tests/unit/__init__.py +0 -0
- package/tests/unit/test_security.py +0 -193
- package/tests/unit/test_setup_wizard.py +0 -35
- package/tests/unit/test_tools.py +0 -320
package/README.md
CHANGED
|
@@ -1,341 +1,267 @@
|
|
|
1
|
-
|
|
2
|
-
|
|
3
|
-
|
|
4
|
-
|
|
5
|
-
|
|
6
|
-
|
|
7
|
-
[![PyPI](https://img.shields.io/pypi/v/ltcai?label=
|
|
1
|
+
<div align="center">
|
|
2
|
+
<img src="https://raw.githubusercontent.com/TaeSooPark-PTS/LatticeAI/main/docs/images/logo.svg" alt="Lattice AI" width="280"/>
|
|
3
|
+
<br/>
|
|
4
|
+
<strong>Your personal AI workspace server — local & cloud, one stack.</strong>
|
|
5
|
+
<br/><br/>
|
|
6
|
+
|
|
7
|
+
[](https://pypi.org/project/ltcai/)
|
|
8
|
+
[](https://pypi.org/project/ltcai/)
|
|
8
9
|
[](https://www.npmjs.com/package/ltcai)
|
|
9
|
-
[](https://marketplace.visualstudio.com/items?itemName=parktaesoo.ltcai)
|
|
10
11
|
[](https://open-vsx.org/extension/parktaesoo/ltcai)
|
|
11
|
-
[](./LICENSE)
|
|
13
|
+
[](https://www.python.org/)
|
|
12
14
|
|
|
13
|
-
|
|
15
|
+
</div>
|
|
14
16
|
|
|
15
|
-
|
|
17
|
+
---
|
|
18
|
+
|
|
19
|
+
## What is Lattice AI?
|
|
16
20
|
|
|
17
|
-
- `
|
|
18
|
-
- `npm`: `ltcai@0.1.22`
|
|
19
|
-
- `VS Code Marketplace`: `parktaesoo.ltcai@0.1.22`
|
|
20
|
-
- `Open VSX`: `parktaesoo.ltcai@0.1.22`
|
|
21
|
+
**Lattice AI** is a self-hosted AI server that unifies local and cloud LLMs into one workspace — web chat, VS Code extension, Telegram bot, and MCP tools, all from a single `pip install`.
|
|
21
22
|
|
|
22
|
-
|
|
23
|
+
- 🖥️ **Web UI** — chat, file upload, admin dashboard, data graph
|
|
24
|
+
- 🧩 **VS Code / Cursor extension** — edit, explain, generate commands inline
|
|
25
|
+
- 📱 **Telegram bot** — access your AI from anywhere
|
|
26
|
+
- 🔌 **MCP server** — use Lattice tools inside Claude Desktop / Cursor
|
|
27
|
+
- 🔒 **Zero telemetry** — all data stays in `~/.ltcai/` on your machine
|
|
23
28
|
|
|
24
|
-
|
|
25
|
-
- **로컬 우선 + 클라우드 선택**: Apple Silicon MLX 로컬 모델과 OpenAI 호환 클라우드 모델을 같은 UX로 다룹니다.
|
|
26
|
-
- **실전형 에이전트 워크플로**: 파일 편집, grep, todo, 터미널 도구를 묶어 멀티스텝 작업을 수행합니다.
|
|
29
|
+
---
|
|
27
30
|
|
|
28
|
-
|
|
31
|
+
## 📸 Screenshots
|
|
29
32
|
|
|
30
|
-
|
|
31
|
-
|
|
32
|
-
|
|
33
|
-
-
|
|
34
|
-
|
|
33
|
+
<table>
|
|
34
|
+
<tr>
|
|
35
|
+
<td width="33%"><b>Chat UI</b><br/><img src="https://raw.githubusercontent.com/TaeSooPark-PTS/LatticeAI/main/docs/images/screenshot-chat.png" alt="Lattice AI Chat" width="100%"/></td>
|
|
36
|
+
<td width="33%"><b>Admin Dashboard</b><br/><img src="https://raw.githubusercontent.com/TaeSooPark-PTS/LatticeAI/main/docs/images/screenshot-admin.png" alt="Admin Dashboard" width="100%"/></td>
|
|
37
|
+
<td width="33%"><b>Data Graph (Graph RAG)</b><br/><img src="https://raw.githubusercontent.com/TaeSooPark-PTS/LatticeAI/main/docs/images/screenshot-graph.png" alt="Knowledge Graph" width="100%"/></td>
|
|
38
|
+
</tr>
|
|
39
|
+
</table>
|
|
35
40
|
|
|
36
41
|
---
|
|
37
42
|
|
|
38
|
-
##
|
|
43
|
+
## ⚡ Quick Start (30 seconds)
|
|
44
|
+
|
|
45
|
+
**Python / PyPI**
|
|
39
46
|
|
|
40
47
|
```bash
|
|
41
|
-
#
|
|
48
|
+
# Install (cloud models)
|
|
42
49
|
pip install ltcai
|
|
43
50
|
|
|
44
|
-
#
|
|
51
|
+
# Install (+ Apple Silicon local models)
|
|
45
52
|
pip install "ltcai[local]"
|
|
46
53
|
|
|
47
|
-
#
|
|
48
|
-
|
|
54
|
+
# Verify environment
|
|
55
|
+
LTCAI doctor
|
|
49
56
|
|
|
50
|
-
#
|
|
57
|
+
# Start server
|
|
51
58
|
LTCAI
|
|
52
59
|
# → http://localhost:4825
|
|
53
|
-
|
|
54
|
-
# 외부에서 접속 가능하게 실행 (Cloudflare 터널 자동 개설)
|
|
55
|
-
LTCAI --tunnel
|
|
56
|
-
# → http://localhost:4825
|
|
57
|
-
# → https://xxxx.trycloudflare.com ← 어디서든 접속 가능한 공개 URL
|
|
58
60
|
```
|
|
59
61
|
|
|
60
|
-
|
|
61
|
-
- cloudflared가 없으면 자동 다운로드 (계정 불필요)
|
|
62
|
-
- 서버를 `0.0.0.0`에 바인딩하고 Cloudflare 무료 터널로 HTTPS 공개 URL 발급
|
|
63
|
-
- `LATTICEAI_TELEGRAM_BOT_TOKEN` + `LATTICEAI_TELEGRAM_CHAT_ID` 환경변수가 있으면 시작 시 Telegram으로 URL 자동 전송
|
|
64
|
-
- 서버 종료 시 터널도 함께 종료
|
|
62
|
+
**Node / npm**
|
|
65
63
|
|
|
66
|
-
|
|
67
|
-
|
|
68
|
-
|
|
69
|
-
|
|
70
|
-
[OK] Python 3.11+: 3.11.9
|
|
71
|
-
[OK] FastAPI: required server dependency
|
|
72
|
-
[OK] Uvicorn: required server dependency
|
|
73
|
-
[OK] OpenAI SDK: required for cloud providers
|
|
74
|
-
[OK] MLX: required for Apple Silicon local models
|
|
75
|
-
[OK] MLX-LM: required for local text models
|
|
76
|
-
[OK] MLX-VLM: required for Gemma/VLM models
|
|
77
|
-
[OPTIONAL] Ollama binary: optional local-server engine
|
|
78
|
-
[OK] Data dir: /Users/you/.ltcai
|
|
79
|
-
[OK] Static UI: /path/to/static
|
|
80
|
-
[INFO] Cloud keys configured: OPENAI_API_KEY
|
|
64
|
+
```bash
|
|
65
|
+
npm install -g ltcai
|
|
66
|
+
LTCAI doctor
|
|
67
|
+
LTCAI
|
|
81
68
|
```
|
|
82
69
|
|
|
83
|
-
|
|
70
|
+
**VS Code / Cursor**
|
|
84
71
|
|
|
85
|
-
|
|
72
|
+
1. Install **Lattice AI** from [VS Code Marketplace](https://marketplace.visualstudio.com/items?itemName=parktaesoo.ltcai) or [Open VSX](https://open-vsx.org/extension/parktaesoo/ltcai)
|
|
73
|
+
2. Start the local server with `LTCAI`
|
|
74
|
+
3. Run `Lattice AI: Open Chat` (`Cmd+Shift+A`) in your editor
|
|
86
75
|
|
|
87
|
-
|
|
76
|
+
**First run:** open `http://localhost:4825` → sign up → first account auto-becomes admin → pick a model → start chatting.
|
|
88
77
|
|
|
89
|
-
|
|
90
|
-
2. **회원가입** → 첫 번째 계정이 자동으로 admin
|
|
91
|
-
3. 상단 모델 드롭다운 → 모델 선택 → 채팅 시작
|
|
92
|
-
|
|
93
|
-
클라우드 모델 사용 시 API 키를 먼저 설정하거나, 어드민 패널에서 입력합니다:
|
|
78
|
+
**Public HTTPS tunnel (Cloudflare, no account needed):**
|
|
94
79
|
|
|
95
80
|
```bash
|
|
96
|
-
|
|
81
|
+
LTCAI --tunnel
|
|
82
|
+
# → https://xxxx.trycloudflare.com
|
|
97
83
|
```
|
|
98
84
|
|
|
99
|
-
|
|
85
|
+
---
|
|
100
86
|
|
|
101
|
-
|
|
102
|
-
2. `Cmd+Shift+A` → Lattice AI 채팅 패널 열기
|
|
103
|
-
3. 기본적으로 `http://localhost:4825` 에 자동 연결됩니다
|
|
87
|
+
## 🆚 Why Lattice AI?
|
|
104
88
|
|
|
105
|
-
|
|
106
|
-
|--------|------|
|
|
107
|
-
| `Cmd+Shift+A` | 채팅 패널 열기 |
|
|
108
|
-
| `Cmd+Shift+E` | 선택 코드 편집 |
|
|
109
|
-
| `Cmd+Shift+M` | 모델 로드 / 전환 |
|
|
110
|
-
| 우클릭 메뉴 | Explain / Edit / Knowledge Garden 저장 |
|
|
89
|
+
Comparison is based on public product behavior as of 2026-05.
|
|
111
90
|
|
|
112
|
-
|
|
91
|
+
| | Lattice AI | Open WebUI | Continue.dev | GitHub Copilot |
|
|
92
|
+
|---|:---:|:---:|:---:|:---:|
|
|
93
|
+
| Local model (offline, Apple Silicon) | ✅ | ✅ | ✅ | ❌ |
|
|
94
|
+
| Cloud models (OpenAI, Groq…) | ✅ | ✅ | ✅ | ✅ |
|
|
95
|
+
| VS Code extension | ✅ | ❌ | ✅ | ✅ |
|
|
96
|
+
| Telegram bot | ✅ | ❌ | ❌ | ❌ |
|
|
97
|
+
| Graph RAG (auto knowledge graph) | ✅ | ❌ | ❌ | ❌ |
|
|
98
|
+
| MCP registry & install | ✅ | ❌ | ✅ | ❌ |
|
|
99
|
+
| Admin dashboard + audit log | ✅ | ✅ | ❌ | ❌ |
|
|
100
|
+
| Self-hosted, zero telemetry | ✅ | ✅ | ✅ | ❌ |
|
|
101
|
+
| One-command public tunnel | ✅ | ❌ | ❌ | ❌ |
|
|
102
|
+
| Free | ✅ | ✅ | ✅ | ❌ |
|
|
113
103
|
|
|
114
|
-
|
|
104
|
+
---
|
|
115
105
|
|
|
116
|
-
|
|
117
|
-
LATTICEAI_TELEGRAM_BOT_TOKEN=your-token LTCAI
|
|
118
|
-
```
|
|
106
|
+
## 🧠 Supported Models
|
|
119
107
|
|
|
120
|
-
|
|
108
|
+
**Local — Apple Silicon only (MLX):**
|
|
109
|
+
|
|
110
|
+
| Model | Best for | Size |
|
|
111
|
+
|-------|----------|------|
|
|
112
|
+
| `mlx-community/gemma-4-26b-a4b-it-4bit` | General / coding | ~14 GB |
|
|
113
|
+
| `mlx-community/Qwen2.5-Coder-32B-Instruct-4bit` | Coding | ~18 GB |
|
|
114
|
+
| `mlx-community/DeepSeek-R1-0528-4bit` | Reasoning | ~38 GB |
|
|
115
|
+
| `mlx-community/Phi-4-4bit` | Coding (fast) | ~8 GB |
|
|
116
|
+
|
|
117
|
+
**Cloud (any platform):**
|
|
118
|
+
OpenAI · Groq · Together · OpenRouter · any OpenAI-compatible endpoint
|
|
121
119
|
|
|
122
120
|
---
|
|
123
121
|
|
|
124
|
-
##
|
|
125
|
-
|
|
126
|
-
|
|
|
127
|
-
|
|
128
|
-
|
|
|
129
|
-
| **Auto Setup Wizard** |
|
|
130
|
-
| **VS Code / Cursor
|
|
131
|
-
| **Telegram
|
|
132
|
-
| **MCP
|
|
133
|
-
| **MCP
|
|
134
|
-
| **Skills
|
|
135
|
-
|
|
|
136
|
-
| **
|
|
137
|
-
|
|
|
138
|
-
| **
|
|
139
|
-
|
|
|
140
|
-
| **
|
|
122
|
+
## ✨ Features
|
|
123
|
+
|
|
124
|
+
| Feature | Description |
|
|
125
|
+
|---------|-------------|
|
|
126
|
+
| **Web UI** | Responsive chat + admin panel + graph visualisation |
|
|
127
|
+
| **Auto Setup Wizard** | Detects → downloads → installs → verifies → repairs dependencies |
|
|
128
|
+
| **VS Code / Cursor** | Chat panel, Edit Selection, Explain, Generate command |
|
|
129
|
+
| **Telegram bot** | Local AI mirror + cloud Codex bot |
|
|
130
|
+
| **MCP server** | Use Lattice tools in Claude Desktop / Cursor |
|
|
131
|
+
| **MCP registry** | One-click install from registry.modelcontextprotocol.io |
|
|
132
|
+
| **Skills marketplace** | 77 official skills (Anthropic + Adobe · Airtable · Auth0 · Pydantic) |
|
|
133
|
+
| **Plugin directory** | Browse 149 open-source plugins |
|
|
134
|
+
| **Graph RAG** | Chat & docs auto-indexed as SQLite knowledge graph |
|
|
135
|
+
| **Multi-step agent** | File edit/create, grep, todo, terminal (25 steps) |
|
|
136
|
+
| **Multi-LLM pipeline** | Plan → Execute → Review with different models |
|
|
137
|
+
| **Human-in-the-loop** | Approve agent plan before execution |
|
|
138
|
+
| **Audit dashboard** | Per-user AI usage, sensitive data detection, event log |
|
|
139
|
+
| **PWA** | Install on iPad / Android home screen |
|
|
141
140
|
| **SSO** | Entra ID / Okta OIDC |
|
|
142
141
|
|
|
143
142
|
---
|
|
144
143
|
|
|
145
|
-
##
|
|
144
|
+
## 🖥️ Platform Support
|
|
146
145
|
|
|
147
|
-
|
|
|
148
|
-
|
|
149
|
-
|
|
|
150
|
-
| VS Code / Cursor
|
|
151
|
-
| Telegram
|
|
152
|
-
| MLX
|
|
153
|
-
| Ollama /
|
|
146
|
+
| Feature | macOS Apple Silicon | macOS Intel / Windows / Linux |
|
|
147
|
+
|---------|:---:|:---:|
|
|
148
|
+
| Web UI + cloud models | ✅ | ✅ |
|
|
149
|
+
| VS Code / Cursor extension | ✅ | ✅ |
|
|
150
|
+
| Telegram bot | ✅ | ✅ |
|
|
151
|
+
| MLX local models | ✅ | ❌ |
|
|
152
|
+
| Ollama / LM Studio / vLLM | ✅ | ✅ |
|
|
154
153
|
|
|
155
154
|
---
|
|
156
155
|
|
|
157
|
-
##
|
|
156
|
+
## 🛠️ Setup & Usage
|
|
157
|
+
|
|
158
|
+
### Install & run
|
|
158
159
|
|
|
159
160
|
```bash
|
|
161
|
+
# Verify everything is ready
|
|
162
|
+
LTCAI doctor
|
|
163
|
+
|
|
164
|
+
# Run with cloud API key
|
|
165
|
+
OPENAI_API_KEY=sk-... LTCAI
|
|
166
|
+
|
|
167
|
+
# Run with local MLX model (Apple Silicon)
|
|
160
168
|
LATTICEAI_MODE=local \
|
|
161
169
|
LATTICEAI_LOCAL_MODEL=mlx-community/gemma-4-26b-a4b-it-4bit \
|
|
162
170
|
LTCAI
|
|
163
171
|
```
|
|
164
172
|
|
|
165
|
-
###
|
|
173
|
+
### VS Code extension
|
|
166
174
|
|
|
167
|
-
|
|
168
|
-
|
|
169
|
-
| `mlx-community/gemma-4-26b-a4b-it-4bit` | 범용/코딩 | ~14GB | ⭐⭐⭐⭐⭐ |
|
|
170
|
-
| `mlx-community/Qwen2.5-Coder-32B-Instruct-4bit` | 코딩 | ~18GB | ⭐⭐⭐⭐⭐ |
|
|
171
|
-
| `mlx-community/Qwen2.5-Coder-14B-Instruct-4bit` | 코딩 | ~8GB | ⭐⭐⭐⭐ |
|
|
172
|
-
| `mlx-community/DeepSeek-R1-0528-4bit` | 추론 | ~38GB | ⭐⭐⭐⭐ |
|
|
173
|
-
| `mlx-community/Phi-4-4bit` | 코딩 | ~8GB | ⭐⭐⭐⭐ |
|
|
175
|
+
1. VS Code → Extensions → search `ltcai` → Install
|
|
176
|
+
2. `Cmd+Shift+A` — open chat panel (auto-connects to `localhost:4825`)
|
|
174
177
|
|
|
175
|
-
|
|
178
|
+
| Shortcut | Action |
|
|
179
|
+
|----------|--------|
|
|
180
|
+
| `Cmd+Shift+A` | Open chat |
|
|
181
|
+
| `Cmd+Shift+E` | Edit selected code |
|
|
182
|
+
| `Cmd+Shift+M` | Load / switch model |
|
|
183
|
+
| Right-click | Explain / Save to Knowledge Garden |
|
|
176
184
|
|
|
177
|
-
###
|
|
185
|
+
### Telegram bot
|
|
178
186
|
|
|
179
187
|
```bash
|
|
180
|
-
|
|
181
|
-
-H "Content-Type: application/json" \
|
|
182
|
-
-d '{"model_id": "mlx-community/Qwen2.5-Coder-14B-Instruct-4bit"}'
|
|
188
|
+
LATTICEAI_TELEGRAM_BOT_TOKEN=your-token LTCAI
|
|
183
189
|
```
|
|
184
190
|
|
|
185
|
-
|
|
186
|
-
|
|
187
|
-
## 퍼블릭 모드 (클라우드 서버)
|
|
188
|
-
|
|
189
|
-
Render, Fly.io, Railway, VPS 등에서 운영할 때:
|
|
191
|
+
### Public server (Render / Fly.io / Docker)
|
|
190
192
|
|
|
191
193
|
```bash
|
|
192
194
|
LATTICEAI_MODE=public \
|
|
193
|
-
LATTICEAI_ALLOW_LOCAL_MODELS=false \
|
|
194
195
|
LATTICEAI_PUBLIC_MODEL=openai:gpt-4o-mini \
|
|
195
196
|
OPENAI_API_KEY=sk-... \
|
|
196
|
-
LATTICEAI_INVITE_CODE=my-secret
|
|
197
|
+
LATTICEAI_INVITE_CODE=my-secret \
|
|
197
198
|
LTCAI
|
|
198
199
|
```
|
|
199
200
|
|
|
200
|
-
자세한 내용은 [docs/public-deploy.md](docs/public-deploy.md)를 참고하세요.
|
|
201
|
-
|
|
202
|
-
### Docker
|
|
203
|
-
|
|
204
201
|
```bash
|
|
202
|
+
# Docker
|
|
205
203
|
docker build -t lattice-ai .
|
|
206
204
|
docker run --rm -p 4825:4825 \
|
|
207
205
|
-e OPENAI_API_KEY="$OPENAI_API_KEY" \
|
|
208
|
-
-e LATTICEAI_INVITE_CODE="my-secret
|
|
206
|
+
-e LATTICEAI_INVITE_CODE="my-secret" \
|
|
209
207
|
-v "$PWD/.data:/data" \
|
|
210
208
|
lattice-ai
|
|
211
209
|
```
|
|
212
210
|
|
|
213
211
|
---
|
|
214
212
|
|
|
215
|
-
##
|
|
213
|
+
## 🔒 Security
|
|
216
214
|
|
|
217
|
-
|
|
218
|
-
|
|
219
|
-
|
|
220
|
-
|
|
221
|
-
|
|
222
|
-
|
|
223
|
-
|
|
224
|
-
|
|
225
|
-
|
|
226
|
-
|
|
227
|
-
- **텔레메트리**: 없음. 모든 데이터는 로컬(`~/.ltcai/`)에만 저장됩니다.
|
|
215
|
+
| Property | Detail |
|
|
216
|
+
|----------|--------|
|
|
217
|
+
| Binding | Default `127.0.0.1:4825` — local only |
|
|
218
|
+
| Auth | Session required when network-exposed or public mode |
|
|
219
|
+
| Cookies | `HttpOnly + SameSite=Lax` — no localStorage token |
|
|
220
|
+
| Local file access | Approval-token gated (path + user + action scope) |
|
|
221
|
+
| CORS | Localhost only by default; add origins via `LATTICEAI_CORS_ALLOWED_ORIGINS` |
|
|
222
|
+
| File upload | Magic-number signature check (blocks extension spoofing) |
|
|
223
|
+
| Rate limits | `/chat` 30/min · `/agent` 6/min · `/upload` 12/min per user |
|
|
224
|
+
| Telemetry | None — all data in `~/.ltcai/` |
|
|
228
225
|
|
|
229
|
-
|
|
226
|
+
Report vulnerabilities: [SECURITY.md](SECURITY.md)
|
|
230
227
|
|
|
231
228
|
---
|
|
232
229
|
|
|
233
|
-
##
|
|
234
|
-
|
|
235
|
-
|
|
|
236
|
-
|
|
237
|
-
|
|
|
238
|
-
|
|
|
239
|
-
|
|
|
240
|
-
|
|
|
241
|
-
|
|
|
242
|
-
|
|
|
243
|
-
|
|
|
244
|
-
|
|
|
230
|
+
## 🗂️ API Reference
|
|
231
|
+
|
|
232
|
+
| Method | Path | Description |
|
|
233
|
+
|--------|------|-------------|
|
|
234
|
+
| GET | `/health` | Server status & current model |
|
|
235
|
+
| GET | `/models` | Model list + load state |
|
|
236
|
+
| POST | `/models/load` | Load a model |
|
|
237
|
+
| POST | `/chat` | Chat (`stream=true/false`) |
|
|
238
|
+
| POST | `/agent` | Multi-step file agent |
|
|
239
|
+
| GET | `/mcp/installed` | Installed MCP servers |
|
|
240
|
+
| POST | `/mcp/install` | Install MCP server |
|
|
241
|
+
| GET | `/skills/marketplace` | Skills marketplace |
|
|
242
|
+
| POST | `/skills/install` | Install a skill |
|
|
243
|
+
| GET | `/plugins/directory` | Plugin directory |
|
|
244
|
+
| GET | `/permissions/pending` | Pending file-access approvals (admin) |
|
|
245
|
+
| POST | `/permissions/approve/{token}` | Approve file access (admin) |
|
|
246
|
+
|
|
247
|
+
Full reference: [docs/mcp-tools.md](docs/mcp-tools.md)
|
|
245
248
|
|
|
246
249
|
---
|
|
247
250
|
|
|
248
|
-
##
|
|
249
|
-
|
|
250
|
-
| 마켓플레이스 | 링크 |
|
|
251
|
-
|---|---|
|
|
252
|
-
| VS Code / Cursor | [marketplace.visualstudio.com](https://marketplace.visualstudio.com/items?itemName=parktaesoo.ltcai) |
|
|
253
|
-
| Antigravity / VSCodium | [open-vsx.org](https://open-vsx.org/extension/parktaesoo/ltcai) |
|
|
254
|
-
|
|
255
|
-
### 릴리스 체크
|
|
256
|
-
|
|
257
|
-
`0.1.22` 릴리스는 아래 네 채널을 동일 버전으로 맞춥니다.
|
|
258
|
-
|
|
259
|
-
- `npm`
|
|
260
|
-
- `PyPI`
|
|
261
|
-
- `VS Code Marketplace`
|
|
262
|
-
- `Open VSX`
|
|
251
|
+
## 🔧 Troubleshooting
|
|
263
252
|
|
|
264
|
-
|
|
265
|
-
|
|
266
|
-
|
|
267
|
-
|
|
268
|
-
|
|
269
|
-
|
|
270
|
-
|
|
271
|
-
|
|
272
|
-
---
|
|
273
|
-
|
|
274
|
-
## 어드민 패널
|
|
275
|
-
|
|
276
|
-
`http://localhost:4825/admin` — 관리자 계정으로 로그인 후 접근
|
|
277
|
-
|
|
278
|
-
- 사용자 목록 및 역할 관리 (admin / user)
|
|
279
|
-
- 대시보드 (메모리, 모델, 시스템 상태, 활동 차트)
|
|
280
|
-
- 초대 링크 생성 및 복사
|
|
281
|
-
|
|
282
|
-
> 첫 번째로 가입한 계정이 자동으로 admin입니다.
|
|
283
|
-
> 환경변수로 고정: `LATTICEAI_ADMIN_EMAILS=you@example.com`
|
|
284
|
-
|
|
285
|
-
---
|
|
286
|
-
|
|
287
|
-
## P-Reinforce 지식 정원
|
|
288
|
-
|
|
289
|
-
코드/텍스트를 `~/.ltcai-brain/`에 자동 분류 저장합니다.
|
|
290
|
-
|
|
291
|
-
```
|
|
292
|
-
~/.ltcai-brain/
|
|
293
|
-
├── 00_Raw/ # 원시 데이터, 아이디어
|
|
294
|
-
├── 10_Wiki/ # 검증된 개념, 레퍼런스
|
|
295
|
-
├── 20_Skills/ # 코드 스니펫, 프롬프트
|
|
296
|
-
├── 30_Projects/ # 프로젝트 컨텍스트
|
|
297
|
-
└── 40_Log/ # 날짜별 작업 로그
|
|
298
|
-
```
|
|
299
|
-
|
|
300
|
-
에디터에서 텍스트 선택 → 우클릭 → **"Save to Knowledge Garden"**
|
|
301
|
-
|
|
302
|
-
---
|
|
303
|
-
|
|
304
|
-
## Data Graph / Graph RAG
|
|
305
|
-
|
|
306
|
-
채팅·AI 답변·업로드 문서(PDF/DOCX/XLSX/PPTX/TXT/CSV)를 `~/.ltcai/knowledge_graph.sqlite`에 자동 저장합니다.
|
|
307
|
-
|
|
308
|
-
- 시각화: `http://localhost:4825/graph`
|
|
309
|
-
- 검색 및 RAG 컨텍스트 자동 주입
|
|
310
|
-
|
|
311
|
-
---
|
|
312
|
-
|
|
313
|
-
## API 엔드포인트
|
|
314
|
-
|
|
315
|
-
| Method | Path | 설명 |
|
|
316
|
-
|--------|------|------|
|
|
317
|
-
| GET | `/health` | 서버 상태, 현재 모델 |
|
|
318
|
-
| GET | `/models` | 추천 모델 목록 + 로드 상태 |
|
|
319
|
-
| POST | `/models/load` | 모델 로드 |
|
|
320
|
-
| POST | `/models/switch/{id}` | 활성 모델 전환 |
|
|
321
|
-
| DELETE | `/models/unload/{id}` | 모델 언로드 |
|
|
322
|
-
| POST | `/chat` | 채팅 생성 (`stream=true/false`) |
|
|
323
|
-
| POST | `/agent` | 파일 생성/수정 에이전트 |
|
|
324
|
-
| GET | `/tools/list_dir` | 디렉토리 목록 |
|
|
325
|
-
| POST | `/tools/run_command` | 터미널 명령 실행 |
|
|
326
|
-
| GET | `/mcp/installed` | 설치된 MCP 목록 (로컬 + 원격 레지스트리) |
|
|
327
|
-
| POST | `/mcp/install` | MCP 설치 (bundled / pip / npm / pypi) |
|
|
328
|
-
| POST | `/mcp/registry/refresh` | 원격 MCP 레지스트리 캐시 갱신 |
|
|
329
|
-
| GET | `/skills/marketplace` | Skills 마켓플레이스 (Anthropic + 서드파티, `?category=` `?author=`) |
|
|
330
|
-
| POST | `/skills/install` | Skill 설치 (`{ "plugin": "...", "skill": "..." }`) |
|
|
331
|
-
| GET | `/skills/list` | 로컬 설치 skills 목록 |
|
|
332
|
-
| GET | `/plugins/directory` | 오픈소스 플러그인 디렉터리 149개 (`?q=` `?category=` `?license=`) |
|
|
333
|
-
|
|
334
|
-
자세한 MCP 도구 목록: [docs/mcp-tools.md](docs/mcp-tools.md)
|
|
253
|
+
| Symptom | Cause | Fix |
|
|
254
|
+
|---------|-------|-----|
|
|
255
|
+
| Port 4825 in use | Previous process | `lsof -i :4825` → `kill <PID>` or `--port 4826` |
|
|
256
|
+
| `ModuleNotFoundError: mlx` | MLX not installed | `pip install "ltcai[local]"` (Apple Silicon only) |
|
|
257
|
+
| Python < 3.11 | Version mismatch | Upgrade: `python3 --version` |
|
|
258
|
+
| `LTCAI doctor` OPTIONAL | Optional dep missing | Safe to ignore if feature not needed |
|
|
259
|
+
| No API key warning | Cloud model not set | `OPENAI_API_KEY=sk-... LTCAI` or set in admin panel |
|
|
260
|
+
| Can't reach from iPad | Default bind 127.0.0.1 | `LATTICEAI_HOST=0.0.0.0 LTCAI` or use `--tunnel` |
|
|
335
261
|
|
|
336
262
|
---
|
|
337
263
|
|
|
338
|
-
##
|
|
264
|
+
## 🚀 Auto-start (Mac)
|
|
339
265
|
|
|
340
266
|
```bash
|
|
341
267
|
cat > ~/Library/LaunchAgents/com.ltcai.plist << 'EOF'
|
|
@@ -344,9 +270,7 @@ cat > ~/Library/LaunchAgents/com.ltcai.plist << 'EOF'
|
|
|
344
270
|
<plist version="1.0">
|
|
345
271
|
<dict>
|
|
346
272
|
<key>Label</key><string>com.ltcai</string>
|
|
347
|
-
<key>ProgramArguments</key><array>
|
|
348
|
-
<string>/usr/local/bin/LTCAI</string>
|
|
349
|
-
</array>
|
|
273
|
+
<key>ProgramArguments</key><array><string>/usr/local/bin/LTCAI</string></array>
|
|
350
274
|
<key>RunAtLoad</key><true/>
|
|
351
275
|
<key>KeepAlive</key><true/>
|
|
352
276
|
<key>StandardOutPath</key><string>/tmp/ltcai.log</string>
|
|
@@ -357,22 +281,69 @@ EOF
|
|
|
357
281
|
launchctl load ~/Library/LaunchAgents/com.ltcai.plist
|
|
358
282
|
```
|
|
359
283
|
|
|
360
|
-
|
|
284
|
+
Or: `./start_ai.sh` (auto-restart + caffeinate)
|
|
361
285
|
|
|
362
286
|
---
|
|
363
287
|
|
|
364
|
-
##
|
|
288
|
+
## 📦 Distribution
|
|
289
|
+
|
|
290
|
+
| Channel | Link |
|
|
291
|
+
|---------|------|
|
|
292
|
+
| PyPI | [pypi.org/project/ltcai](https://pypi.org/project/ltcai/) |
|
|
293
|
+
| npm | [npmjs.com/package/ltcai](https://www.npmjs.com/package/ltcai) |
|
|
294
|
+
| VS Code Marketplace | [marketplace.visualstudio.com](https://marketplace.visualstudio.com/items?itemName=parktaesoo.ltcai) |
|
|
295
|
+
| Open VSX | [open-vsx.org](https://open-vsx.org/extension/parktaesoo/ltcai) |
|
|
365
296
|
|
|
366
|
-
[
|
|
297
|
+
Current version: **0.1.24** — [Changelog](docs/CHANGELOG.md)
|
|
367
298
|
|
|
368
|
-
|
|
299
|
+
---
|
|
369
300
|
|
|
370
|
-
|
|
301
|
+
## 🤝 Contributing
|
|
302
|
+
|
|
303
|
+
See [CONTRIBUTING.md](CONTRIBUTING.md). All PRs welcome.
|
|
304
|
+
|
|
305
|
+
## 📄 License
|
|
306
|
+
|
|
307
|
+
MIT — [TaeSoo Park](https://github.com/TaeSooPark-PTS)
|
|
308
|
+
|
|
309
|
+
---
|
|
371
310
|
|
|
372
|
-
|
|
311
|
+
<details>
|
|
312
|
+
<summary>한국어 안내 (Korean)</summary>
|
|
373
313
|
|
|
374
|
-
|
|
314
|
+
## 한국어 안내
|
|
315
|
+
|
|
316
|
+
**Lattice AI**는 로컬/클라우드 LLM을 웹 UI · VS Code 확장 · Telegram 봇 · MCP 도구로 하나의 서버에서 운영하는 개인 AI 워크스페이스입니다.
|
|
317
|
+
|
|
318
|
+
### 설치
|
|
319
|
+
|
|
320
|
+
```bash
|
|
321
|
+
pip install ltcai # 클라우드 모델
|
|
322
|
+
pip install "ltcai[local]" # + Apple Silicon MLX 로컬 모델
|
|
323
|
+
LTCAI # 서버 실행 → http://localhost:4825
|
|
324
|
+
LTCAI --tunnel # + Cloudflare 공개 URL 자동 발급
|
|
325
|
+
```
|
|
326
|
+
|
|
327
|
+
### 주요 기능
|
|
328
|
+
|
|
329
|
+
- 웹 UI 채팅 + 어드민 대시보드 + Data Graph 시각화
|
|
330
|
+
- VS Code / Cursor 확장 (`Cmd+Shift+A`)
|
|
331
|
+
- Telegram 봇 연동
|
|
332
|
+
- MCP 레지스트리 & Skills 마켓플레이스
|
|
333
|
+
- Graph RAG — 채팅·문서를 SQLite 지식 그래프로 자동 구조화
|
|
334
|
+
- 멀티 LLM 파이프라인 (Plan → Execute → Review)
|
|
335
|
+
- Human-in-the-loop 에이전트 승인
|
|
336
|
+
- 감사 로그 & 데이터 거버넌스 대시보드
|
|
337
|
+
- 텔레메트리 없음 — 모든 데이터 로컬 저장
|
|
338
|
+
|
|
339
|
+
### 추천 로컬 모델 (M-series Mac)
|
|
340
|
+
|
|
341
|
+
| 모델 | 용도 | 크기 |
|
|
342
|
+
|------|------|------|
|
|
343
|
+
| `mlx-community/gemma-4-26b-a4b-it-4bit` | 범용 | ~14GB |
|
|
344
|
+
| `mlx-community/Qwen2.5-Coder-32B-Instruct-4bit` | 코딩 | ~18GB |
|
|
345
|
+
| `mlx-community/DeepSeek-R1-0528-4bit` | 추론 | ~38GB |
|
|
375
346
|
|
|
376
|
-
|
|
347
|
+
자세한 내용: [docs/CHANGELOG.md](docs/CHANGELOG.md) · [보안](SECURITY.md) · [기여](CONTRIBUTING.md)
|
|
377
348
|
|
|
378
|
-
|
|
349
|
+
</details>
|