aiteam-x 0.9.7 → 0.11.0
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 +21 -0
- package/README.en.md +425 -0
- package/README.md +425 -397
- package/package.json +69 -55
- package/scripts/cli.mjs +186 -171
package/LICENSE
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026 INOSX
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
package/README.en.md
ADDED
|
@@ -0,0 +1,425 @@
|
|
|
1
|
+
<p align="right">
|
|
2
|
+
<a href="README.md">Portugues</a> | <strong>English</strong>
|
|
3
|
+
</p>
|
|
4
|
+
|
|
5
|
+
# AITeam Team Overview
|
|
6
|
+
|
|
7
|
+
**Pixel-art visual dashboard for orchestrating and monitoring BMAD agents in real time.**
|
|
8
|
+
|
|
9
|
+
[](https://www.npmjs.com/package/aiteam-x)
|
|
10
|
+
[](https://nextjs.org/)
|
|
11
|
+
[](https://react.dev/)
|
|
12
|
+
[](https://www.typescriptlang.org/)
|
|
13
|
+
[](https://github.com/bmad-method)
|
|
14
|
+
[](LICENSE)
|
|
15
|
+
|
|
16
|
+
```bash
|
|
17
|
+
npx aiteam-x@latest
|
|
18
|
+
```
|
|
19
|
+
|
|
20
|
+
---
|
|
21
|
+
|
|
22
|
+
## Overview
|
|
23
|
+
|
|
24
|
+
**AITeam Team Overview** is an interactive dashboard that turns AI agent management into a visual, intuitive experience. Inspired by *sim manager*-style simulation games, the panel displays BMAD agents as pixel-art characters in a virtual office, allowing you to:
|
|
25
|
+
|
|
26
|
+
- **Visualize** each agent's status and location in real time
|
|
27
|
+
- **Interact** directly with agents through integrated chat windows
|
|
28
|
+
- **Orchestrate** workflows such as Party Mode and brainstorming sessions
|
|
29
|
+
- **Monitor** usage metrics (messages, tokens in/out) per agent
|
|
30
|
+
- **Persist** context and memories between work sessions with LLM extraction
|
|
31
|
+
|
|
32
|
+
The project uses the **BMAD Method** as the agent framework, natively integrating with the **Cursor Agent CLI** and **Cursor ACP** (Agent Client Protocol) for bidirectional LLM communication.
|
|
33
|
+
|
|
34
|
+
---
|
|
35
|
+
|
|
36
|
+
## Screenshots
|
|
37
|
+
|
|
38
|
+
### Main Dashboard
|
|
39
|
+
|
|
40
|
+
|
|
41
|
+
|
|
42
|
+
*Virtual office with Jarvis Office, Kitchen, and individual workspaces. Bottom bar with avatars and sprint progress (14/14 — 100%).*
|
|
43
|
+
|
|
44
|
+
---
|
|
45
|
+
|
|
46
|
+
### Agent Chat Window
|
|
47
|
+
|
|
48
|
+
|
|
49
|
+
|
|
50
|
+
*Real-time conversation window with BMad Master. Each agent responds with text streaming and Markdown support.*
|
|
51
|
+
|
|
52
|
+
---
|
|
53
|
+
|
|
54
|
+
### Memory Vault
|
|
55
|
+
|
|
56
|
+
|
|
57
|
+
|
|
58
|
+
*Per-agent memory management panel. Categories: decisions, lessons, tasks, projects, handoffs. Real-time BM25 search.*
|
|
59
|
+
|
|
60
|
+
---
|
|
61
|
+
|
|
62
|
+
### Agent Filters
|
|
63
|
+
|
|
64
|
+
|
|
65
|
+
|
|
66
|
+
*Status filters (All / Working) and Module filters (BMGD / BMM / CORE) for segmented agent views.*
|
|
67
|
+
|
|
68
|
+
---
|
|
69
|
+
|
|
70
|
+
## Key Features
|
|
71
|
+
|
|
72
|
+
### Pixel Art Dashboard
|
|
73
|
+
- Shared rooms: **Jarvis Office**, **Kitchen** — with agent activity context
|
|
74
|
+
- Individual workspace grid per agent with module indicators (BMGD / BMM / CORE)
|
|
75
|
+
- Pixel-art avatars with real-time visual status
|
|
76
|
+
- Bottom bar with compact avatars and sprint timeline with granular per-story progress
|
|
77
|
+
- Day/night cycle and ambient effects
|
|
78
|
+
|
|
79
|
+
### Agent Interaction
|
|
80
|
+
- Agent selection (click) or multi-selection (Ctrl+click) for batch commands
|
|
81
|
+
- Resizable chat windows with real-time SSE streaming
|
|
82
|
+
- Animated "thinking" bubble during processing
|
|
83
|
+
- Full Markdown rendering in responses
|
|
84
|
+
- **Copy button**: appears on hover over agent responses for one-click copy
|
|
85
|
+
- **Context menu**: right-click on any agent to see agent-specific MDC commands
|
|
86
|
+
- Stop button to cancel ongoing processing
|
|
87
|
+
- Conversation history persisted across reloads
|
|
88
|
+
- Dynamic z-index system (bring-to-front on click)
|
|
89
|
+
|
|
90
|
+
### Model Selector & Filters
|
|
91
|
+
- **Model Bar**: LLM model selection applied to all agents
|
|
92
|
+
- **Cursor ACP**: dynamic submenu listing all models available in the user's Cursor account (Claude, GPT, Gemini, Grok, etc.)
|
|
93
|
+
- **Direct models**: Composer 2, Claude Opus/Sonnet, GPT-5.x, Gemini, Grok, Kimi
|
|
94
|
+
- **Status Filter**: All / Working — displays real-time counter (e.g. 14/14)
|
|
95
|
+
- **Module Filter**: BMGD / BMM / CORE — segments agents by area
|
|
96
|
+
|
|
97
|
+
### Cursor ACP (Agent Client Protocol)
|
|
98
|
+
- Direct communication with Cursor via JSON-RPC 2.0 over stdio (`agent acp`)
|
|
99
|
+
- Automatic discovery of available models via ACP session `configOptions`
|
|
100
|
+
- Per-session model switching via `session/set_config_option`
|
|
101
|
+
- Persistent ACP process with 30-min session TTL and automatic reconnection
|
|
102
|
+
- Error fallback with actionable messages for the user
|
|
103
|
+
|
|
104
|
+
### BMAD Orchestration
|
|
105
|
+
- Automatic detection of Party Mode and active workflows
|
|
106
|
+
- Agents dynamically move to rooms according to the workflow
|
|
107
|
+
- Session state persisted on disk and synchronized via SSE
|
|
108
|
+
- Per-agent message and token metrics
|
|
109
|
+
|
|
110
|
+
### Persistent Memory System (v2.1)
|
|
111
|
+
- **Memory Vault**: full visual panel with 5 categories (decisions, lessons, tasks, projects, handoffs)
|
|
112
|
+
- **Auto-extract via LLM**: when closing a session, the system analyzes the conversation and automatically extracts structured memories
|
|
113
|
+
- **Badge 🤖 llm**: automatically extracted entries are flagged for 10 minutes with veto option
|
|
114
|
+
- **BM25 Search**: relevance-based semantic search within each agent's vault
|
|
115
|
+
- **Context Injection**: relevant memories automatically injected when starting new sessions (2,000-token budget)
|
|
116
|
+
- **Auto-save**: conversations saved every 30s + save on unload via `sendBeacon`
|
|
117
|
+
- **Session Checkpoint**: history preserved even across unexpected restarts
|
|
118
|
+
- **Shared Memory**: `_project.md` injected into all agents as global context
|
|
119
|
+
|
|
120
|
+
### Portability & Installation (v1.0)
|
|
121
|
+
- **Dynamic Discovery**: automatically detects BMAD agents via CSV/YAML — no hardcoding
|
|
122
|
+
- **Cross-platform**: works on Windows, macOS, and Linux
|
|
123
|
+
- **Web Setup Wizard**: `/setup` in the browser — 3 animated phases: machine scan, review, step-by-step configuration
|
|
124
|
+
- **CLI Setup Wizard**: `npm run setup` — interactive terminal alternative (6 steps)
|
|
125
|
+
- **Health Gate**: if not configured, automatically redirects to `/setup`
|
|
126
|
+
- **Flexible Config**: local `aiteam.config.json` + `AITEAM_BMAD_ROOT` for external/CI instances
|
|
127
|
+
|
|
128
|
+
---
|
|
129
|
+
|
|
130
|
+
## Tech Stack
|
|
131
|
+
|
|
132
|
+
| Layer | Technology | Version |
|
|
133
|
+
|-------|-----------|---------|
|
|
134
|
+
| Framework | Next.js (App Router) | 15.x |
|
|
135
|
+
| UI | React | 19.x |
|
|
136
|
+
| Language | TypeScript | 5.x |
|
|
137
|
+
| Styling | CSS custom (pixel art design system) | — |
|
|
138
|
+
| Font | VT323 (Google Fonts) | — |
|
|
139
|
+
| Backend | Next.js API Routes + SSE | — |
|
|
140
|
+
| Search | MiniSearch (BM25) | — |
|
|
141
|
+
| Agents | Cursor Agent CLI + ACP | latest |
|
|
142
|
+
| Protocol | ACP (JSON-RPC 2.0 / stdio) | v1 |
|
|
143
|
+
| Parsing | yaml (npm) | 2.8.x |
|
|
144
|
+
| Agent Framework | BMAD Method | — |
|
|
145
|
+
|
|
146
|
+
---
|
|
147
|
+
|
|
148
|
+
## Quick Start
|
|
149
|
+
|
|
150
|
+
### Prerequisites
|
|
151
|
+
|
|
152
|
+
| Requirement | Required | Notes |
|
|
153
|
+
|-------------|----------|-------|
|
|
154
|
+
| Node.js >= 18 | Yes | `node --version` |
|
|
155
|
+
| BMAD Framework | Yes | See [Installation Guide](docs/installation-guide.md) |
|
|
156
|
+
| Cursor IDE | Yes | Generates the Cursor Agent CLI on first launch |
|
|
157
|
+
| Git | Yes | — |
|
|
158
|
+
|
|
159
|
+
> **Supported platforms:** Windows 10/11, macOS, Linux
|
|
160
|
+
|
|
161
|
+
### Installation
|
|
162
|
+
|
|
163
|
+
**Option 1 — One command (recommended):**
|
|
164
|
+
|
|
165
|
+
```bash
|
|
166
|
+
npx aiteam-x@latest
|
|
167
|
+
```
|
|
168
|
+
|
|
169
|
+
Creates the project in the current directory (must be empty). Or specify a folder name:
|
|
170
|
+
|
|
171
|
+
```bash
|
|
172
|
+
npx aiteam-x@latest my-project
|
|
173
|
+
```
|
|
174
|
+
|
|
175
|
+
The command downloads the template, installs dependencies, runs the setup wizard, and starts the server automatically.
|
|
176
|
+
|
|
177
|
+
**Option 2 — Manual installation:**
|
|
178
|
+
|
|
179
|
+
```bash
|
|
180
|
+
# 1. Clone the repository
|
|
181
|
+
git clone https://github.com/INOSX/AITeam.git
|
|
182
|
+
cd AITeam
|
|
183
|
+
|
|
184
|
+
# 2. Install dependencies
|
|
185
|
+
npm install
|
|
186
|
+
|
|
187
|
+
# 3. Run the setup wizard
|
|
188
|
+
npm run setup
|
|
189
|
+
|
|
190
|
+
# 4. Start the development server
|
|
191
|
+
npm run dev
|
|
192
|
+
```
|
|
193
|
+
|
|
194
|
+
The dashboard will be available at **http://localhost:3000**.
|
|
195
|
+
|
|
196
|
+
> If setup hasn't been completed, the system automatically redirects to `http://localhost:3000/setup` — an animated wizard that detects the environment, shows what's ready, and lets you configure everything through the browser.
|
|
197
|
+
|
|
198
|
+
> **CLI alternative:** `npm run setup` runs the interactive wizard in the terminal.
|
|
199
|
+
|
|
200
|
+
### Available Scripts
|
|
201
|
+
|
|
202
|
+
| Command | Description |
|
|
203
|
+
|---------|-------------|
|
|
204
|
+
| `npm run setup` | **Interactive setup wizard** (first install) |
|
|
205
|
+
| `npm run dev` | Start the development server (port 3000) |
|
|
206
|
+
| `npm run build` | Generate the production build |
|
|
207
|
+
| `npm run start` | Start the production server |
|
|
208
|
+
| `npm run lint` | Run the linter (ESLint) |
|
|
209
|
+
|
|
210
|
+
### Utility Scripts
|
|
211
|
+
|
|
212
|
+
| Script | Description |
|
|
213
|
+
|--------|-------------|
|
|
214
|
+
| `node scripts/extract-memories.mjs --all` | Extract memories from historical conversations via LLM |
|
|
215
|
+
| `node scripts/extract-memories.mjs --agent bmad-master` | Extract memories for a specific agent |
|
|
216
|
+
| `node scripts/extract-memories.mjs --all --dry-run` | Simulate extraction without saving |
|
|
217
|
+
| `node scripts/import-conversations.mjs` | Import conversation history into the vault |
|
|
218
|
+
|
|
219
|
+
> All scripts require the server running at `localhost:3000`.
|
|
220
|
+
|
|
221
|
+
---
|
|
222
|
+
|
|
223
|
+
## Project Structure
|
|
224
|
+
|
|
225
|
+
```
|
|
226
|
+
AITeam/
|
|
227
|
+
├── app/ # Next.js App Router
|
|
228
|
+
│ ├── api/ # API Routes (backend)
|
|
229
|
+
│ │ ├── acp/
|
|
230
|
+
│ │ │ └── models/route.ts # GET: available models via ACP
|
|
231
|
+
│ │ ├── agents/
|
|
232
|
+
│ │ │ ├── route.ts # GET: list agents with status
|
|
233
|
+
│ │ │ ├── command/route.ts # POST: send command via SSE (CLI or ACP)
|
|
234
|
+
│ │ │ └── menus/route.ts # GET: agent context menus
|
|
235
|
+
│ │ ├── bridge/route.ts # GET: ACP health check
|
|
236
|
+
│ │ ├── config/route.ts # GET: BMAD configuration
|
|
237
|
+
│ │ ├── memory/
|
|
238
|
+
│ │ │ ├── route.ts # GET/POST: memory and conversations (legacy)
|
|
239
|
+
│ │ │ ├── vault/route.ts # GET/POST/PUT/DELETE: structured vault
|
|
240
|
+
│ │ │ ├── search/route.ts # GET: BM25 search
|
|
241
|
+
│ │ │ ├── checkpoint/route.ts # POST: session checkpoint
|
|
242
|
+
│ │ │ ├── sleep/route.ts # POST: end session + queue LLM extraction
|
|
243
|
+
│ │ │ └── extract/route.ts # POST: manual LLM extraction trigger
|
|
244
|
+
│ │ ├── session/
|
|
245
|
+
│ │ │ ├── route.ts # GET/POST: active session
|
|
246
|
+
│ │ │ └── stream/route.ts # GET: SSE for session changes
|
|
247
|
+
│ │ ├── sprint/route.ts # GET/POST: sprint status
|
|
248
|
+
│ │ └── workflows/route.ts # GET: available workflows
|
|
249
|
+
│ ├── actions/
|
|
250
|
+
│ │ └── setup-save.ts # Server Action: saves aiteam.config.json
|
|
251
|
+
│ ├── setup/
|
|
252
|
+
│ │ ├── page.tsx # Setup page (health gate + Server Component)
|
|
253
|
+
│ │ └── SetupWizard.tsx # Animated 3-phase wizard (Client Component)
|
|
254
|
+
│ ├── globals.css # Pixel art design system (800+ lines)
|
|
255
|
+
│ ├── layout.tsx # Root layout (VT323, pt-BR)
|
|
256
|
+
│ ├── error.tsx # Error boundary
|
|
257
|
+
│ └── page.tsx # Main page (health gate)
|
|
258
|
+
│
|
|
259
|
+
├── components/ # React Components
|
|
260
|
+
│ ├── rooms/ # Virtual office rooms
|
|
261
|
+
│ │ ├── ConferenceRoom.tsx
|
|
262
|
+
│ │ ├── JarvisOffice.tsx
|
|
263
|
+
│ │ └── Kitchen.tsx
|
|
264
|
+
│ ├── AgentContextMenu.tsx # Per-agent context menu
|
|
265
|
+
│ ├── AgentsProvider.tsx # Global Context Provider
|
|
266
|
+
│ ├── CommandPopup.tsx # Multi-agent popup
|
|
267
|
+
│ ├── CopyButton.tsx # Copy button for responses
|
|
268
|
+
│ ├── FilterContext.tsx # Filter context (status + module)
|
|
269
|
+
│ ├── Layout.tsx # Shell (TopBar + ModelBar + BottomBar)
|
|
270
|
+
│ ├── MainContent.tsx # Main orchestrator (~500 lines)
|
|
271
|
+
│ ├── MarkdownRenderer.tsx # Custom Markdown renderer
|
|
272
|
+
│ ├── MemoryVault.tsx # Memory Vault dashboard (modal)
|
|
273
|
+
│ ├── ModelBar.tsx # Model selection bar and filters
|
|
274
|
+
│ ├── PixelAvatar.tsx # Pixel art SVG avatar
|
|
275
|
+
│ ├── SpeechBubbleOverlay.tsx # Chat overlay system
|
|
276
|
+
│ ├── ThinkingBubble.tsx # Animated thinking indicator
|
|
277
|
+
│ ├── WorkspaceCell.tsx # Individual workspace cell
|
|
278
|
+
│ └── WorkspaceGrid.tsx # Workspace grid
|
|
279
|
+
│
|
|
280
|
+
├── lib/ # Business logic and utilities
|
|
281
|
+
│ ├── acp-client.ts # ACP client (JSON-RPC 2.0 over stdio)
|
|
282
|
+
│ ├── acp-manager.ts # ACP session manager
|
|
283
|
+
│ ├── aiteam-config.ts # Config loader (env var > file > defaults)
|
|
284
|
+
│ ├── health.ts # Health check: BMAD, Cursor CLI, config
|
|
285
|
+
│ ├── setup-discovery.ts # Extended discovery for the web wizard
|
|
286
|
+
│ ├── bmad/ # BMAD Integration
|
|
287
|
+
│ │ ├── discovery.ts # Dynamic BMAD scanner + Cursor CLI detection
|
|
288
|
+
│ │ ├── agent-mapping.ts # BmadAgent -> Agent mapping
|
|
289
|
+
│ │ ├── chat-sessions.ts # Chat session persistence
|
|
290
|
+
│ │ ├── parse-agents.ts # agent-manifest.csv parser
|
|
291
|
+
│ │ ├── parse-config.ts # config.yaml parser
|
|
292
|
+
│ │ ├── parse-persona.ts # Persona (.mdc) parser
|
|
293
|
+
│ │ ├── parse-session.ts # Session state + SSE
|
|
294
|
+
│ │ ├── parse-sprint.ts # sprint-status.yaml parser
|
|
295
|
+
│ │ └── parse-workflows.ts # Workflows parser
|
|
296
|
+
│ ├── memory/ # Memory system v2
|
|
297
|
+
│ │ ├── extract.ts # LLM extraction engine
|
|
298
|
+
│ │ ├── inject.ts # Context injection (BM25 + budget)
|
|
299
|
+
│ │ ├── session.ts # Session lifecycle (sleep/recover)
|
|
300
|
+
│ │ ├── types.ts # Memory system types
|
|
301
|
+
│ │ └── vault.ts # Structured vault CRUD
|
|
302
|
+
│ ├── model-config.ts # Available LLM model options
|
|
303
|
+
│ └── types.ts # Global TypeScript types
|
|
304
|
+
│
|
|
305
|
+
├── .memory/ # Memory Bank (persistent, gitignored)
|
|
306
|
+
│ ├── _project.md # Global context — injected into all agents
|
|
307
|
+
│ ├── {agent-id}.md # Individual memory bank per agent
|
|
308
|
+
│ ├── conversations/ # Conversation history (JSON)
|
|
309
|
+
│ └── .vault/ # Structured vault and extraction queue
|
|
310
|
+
│
|
|
311
|
+
├── docs/ # Project documentation
|
|
312
|
+
│ ├── screenshots/ # UI screenshots
|
|
313
|
+
│ ├── installation-guide.md # Full installation guide
|
|
314
|
+
│ ├── architecture.md # System architecture
|
|
315
|
+
│ ├── api-reference.md # Complete API reference
|
|
316
|
+
│ ├── developer-guide.md # Developer guide
|
|
317
|
+
│ ├── user-guide.md # End-user guide
|
|
318
|
+
│ ├── memory-system.md # Memory system technical architecture
|
|
319
|
+
│ ├── memory-system-guide.md # Memory system functional guide
|
|
320
|
+
│ ├── agent-catalog.md # Complete BMAD agent catalog
|
|
321
|
+
│ └── sprint-status.yaml # Current sprint status
|
|
322
|
+
│
|
|
323
|
+
├── scripts/ # Utility scripts
|
|
324
|
+
│ ├── setup.mjs # Interactive setup wizard (6 steps)
|
|
325
|
+
│ ├── extract-memories.mjs # Batch LLM extraction
|
|
326
|
+
│ └── import-conversations.mjs # History import into vault
|
|
327
|
+
│
|
|
328
|
+
├── aiteam.config.example.json # Configuration example (reference)
|
|
329
|
+
├── .cursor/ # Cursor IDE rules and skills
|
|
330
|
+
│ └── rules/bmad/ # BMAD rules (.mdc): agents and workflows
|
|
331
|
+
│
|
|
332
|
+
├── package.json
|
|
333
|
+
├── tsconfig.json
|
|
334
|
+
├── next.config.ts
|
|
335
|
+
└── .gitignore
|
|
336
|
+
```
|
|
337
|
+
|
|
338
|
+
---
|
|
339
|
+
|
|
340
|
+
## Documentation
|
|
341
|
+
|
|
342
|
+
| Document | Level | Description |
|
|
343
|
+
|----------|-------|-------------|
|
|
344
|
+
| [Installation Guide](docs/installation-guide.md) | Beginner | Step-by-step installation, setup wizard, troubleshooting |
|
|
345
|
+
| [Architecture](docs/architecture.md) | Technical/Architectural | Diagrams, data flows, design decisions |
|
|
346
|
+
| [API Reference](docs/api-reference.md) | Technical | Endpoints, parameters, request/response examples |
|
|
347
|
+
| [Developer Guide](docs/developer-guide.md) | Developer | Setup, conventions, how to contribute |
|
|
348
|
+
| [User Guide](docs/user-guide.md) | End User | How to use the dashboard, interact with agents, Memory Vault |
|
|
349
|
+
| [Memory System (technical)](docs/memory-system.md) | Technical/Conceptual | Layered architecture of the memory system |
|
|
350
|
+
| [Memory Guide (functional)](docs/memory-system-guide.md) | End User | How to use the Memory Vault, case studies, best practices |
|
|
351
|
+
| [Agent Catalog](docs/agent-catalog.md) | Reference | All agents, specialties, and how to work with each one |
|
|
352
|
+
|
|
353
|
+
---
|
|
354
|
+
|
|
355
|
+
## BMAD Agents
|
|
356
|
+
|
|
357
|
+
The system automatically discovers agents from `bmad/_cfg/agent-manifest.csv`. The default installation includes 14 agents:
|
|
358
|
+
|
|
359
|
+
| Persona | Role | Module |
|
|
360
|
+
|---------|------|--------|
|
|
361
|
+
| **BMad Master** | Main orchestrator, knowledge custodian | core |
|
|
362
|
+
| **BMad Builder** | BMAD module, agent, and workflow creation | core |
|
|
363
|
+
| **Mary** | Business Analyst — research and business requirements | bmm |
|
|
364
|
+
| **John** | Product Manager — PRD, epics, product management | bmm |
|
|
365
|
+
| **Winston** | Architect — system architecture and tech spec | bmm |
|
|
366
|
+
| **Sally** | UX Designer — experience and interface design | bmm |
|
|
367
|
+
| **Bob** | Scrum Master — sprint planning and code review | bmm |
|
|
368
|
+
| **Amelia** | Developer — story and feature implementation | bmm |
|
|
369
|
+
| **Murat** | Test Architect — testing strategy and design | bmm |
|
|
370
|
+
| **Paige** | Tech Writer — technical documentation and knowledge | bmm |
|
|
371
|
+
| **Cloud Dragonborn** | Game Architect — game architecture and systems | bmgd |
|
|
372
|
+
| **Samus Shepard** | Game Designer — GDD, narrative, and game design | bmgd |
|
|
373
|
+
| **Link Freeman** | Game Developer — game implementation | bmgd |
|
|
374
|
+
| **Max** | Game Scrum Master — agile management for game dev | bmgd |
|
|
375
|
+
|
|
376
|
+
> See the [Agent Catalog](docs/agent-catalog.md) for details on each agent's specialties and commands.
|
|
377
|
+
|
|
378
|
+
---
|
|
379
|
+
|
|
380
|
+
## Configuration
|
|
381
|
+
|
|
382
|
+
AITEAM-X uses a local `aiteam.config.json` file (gitignored) generated by `npm run setup`. For advanced configuration, create a `.env.local`:
|
|
383
|
+
|
|
384
|
+
```env
|
|
385
|
+
# Alternate path to the BMAD installation
|
|
386
|
+
# Useful for: BMAD in another directory, CI/CD, multiple projects
|
|
387
|
+
AITEAM_BMAD_ROOT=/path/to/your/bmad-project
|
|
388
|
+
|
|
389
|
+
# Overrides the memory directory (useful for testing)
|
|
390
|
+
MEMORY_DIR_OVERRIDE=/custom/path/.memory
|
|
391
|
+
```
|
|
392
|
+
|
|
393
|
+
**Configuration priority for `bmadRoot`:**
|
|
394
|
+
```
|
|
395
|
+
AITEAM_BMAD_ROOT (env var) > aiteam.config.json > "./" (default)
|
|
396
|
+
```
|
|
397
|
+
|
|
398
|
+
See the [Installation Guide](docs/installation-guide.md) for all configuration scenarios.
|
|
399
|
+
|
|
400
|
+
---
|
|
401
|
+
|
|
402
|
+
## Contributing
|
|
403
|
+
|
|
404
|
+
1. Create a branch from `develop`
|
|
405
|
+
2. Make your changes following project conventions
|
|
406
|
+
3. Ensure `npm run lint` and `npm run build` pass
|
|
407
|
+
4. Open a Pull Request to `develop`
|
|
408
|
+
|
|
409
|
+
See the [Developer Guide](docs/developer-guide.md) for details.
|
|
410
|
+
|
|
411
|
+
---
|
|
412
|
+
|
|
413
|
+
## License
|
|
414
|
+
|
|
415
|
+
This project is distributed under the MIT License. See [LICENSE](LICENSE) for details.
|
|
416
|
+
|
|
417
|
+
---
|
|
418
|
+
|
|
419
|
+
## Author
|
|
420
|
+
|
|
421
|
+
**Mario Mayerle (MaMFLux)** — [INOSX](https://inosx.com)
|
|
422
|
+
|
|
423
|
+
---
|
|
424
|
+
|
|
425
|
+
*Built with the BMAD Method — where AI agents work as a team.*
|