@agentic-robotics/mcp 0.1.3

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.
Files changed (16) hide show
  1. package/README.md +300 -0
  2. package/package.json +51 -0
  3. package/src/benchmark.ts +434 -0
  4. package/src/cli.ts +89 -0
  5. package/src/enhanced-memory.ts +437 -0
  6. package/src/enhanced-server.ts +523 -0
  7. package/src/flow-orchestrator.ts +415 -0
  8. package/src/hybrid-benchmark.ts +291 -0
  9. package/src/hybrid-memory.ts +401 -0
  10. package/src/index.ts +10 -0
  11. package/src/interface.ts +118 -0
  12. package/src/memory.ts +196 -0
  13. package/src/optimized-benchmark.ts +302 -0
  14. package/src/optimized-memory.ts +271 -0
  15. package/src/server.ts +213 -0
  16. package/tsconfig.json +20 -0
package/README.md ADDED
@@ -0,0 +1,300 @@
1
+ # @agentic-robotics/mcp
2
+
3
+ [![npm version](https://img.shields.io/npm/v/@agentic-robotics/mcp.svg)](https://www.npmjs.com/package/@agentic-robotics/mcp)
4
+ [![Downloads](https://img.shields.io/npm/dm/@agentic-robotics/mcp.svg)](https://www.npmjs.com/package/@agentic-robotics/mcp)
5
+ [![License](https://img.shields.io/npm/l/@agentic-robotics/mcp.svg)](https://github.com/ruvnet/vibecast)
6
+
7
+ **Model Context Protocol (MCP) server** for agentic robotics with AgentDB memory and agentic-flow orchestration.
8
+
9
+ ## Features
10
+
11
+ - 🤖 **21 MCP Tools** - Complete AI-robot integration toolkit
12
+ - 🧠 **AgentDB Memory** - 13,000x faster than CLI (reflexion, skills, causal reasoning)
13
+ - 🌊 **Agentic Flow** - 66 AI agents + 213 MCP tools orchestration
14
+ - ⚡ **High Performance** - 5,725 ops/sec storage throughput
15
+ - 🔄 **Multi-Robot Swarm** - Coordinate multiple robots intelligently
16
+ - 📊 **Performance Tracking** - Built-in metrics and benchmarking
17
+ - 🎯 **Production Ready** - Optimized hybrid SQL implementation
18
+
19
+ ## Installation
20
+
21
+ ```bash
22
+ npm install -g @agentic-robotics/mcp
23
+ ```
24
+
25
+ Or add to your project:
26
+
27
+ ```bash
28
+ npm install @agentic-robotics/mcp
29
+ ```
30
+
31
+ ## Quick Start
32
+
33
+ ### Start MCP Server
34
+
35
+ ```bash
36
+ agentic-robotics-mcp
37
+ ```
38
+
39
+ ### Use with Claude Desktop
40
+
41
+ Add to your Claude config (`~/Library/Application Support/Claude/claude_desktop_config.json`):
42
+
43
+ ```json
44
+ {
45
+ "mcpServers": {
46
+ "agentic-robotics": {
47
+ "command": "agentic-robotics-mcp",
48
+ "args": []
49
+ }
50
+ }
51
+ }
52
+ ```
53
+
54
+ ## MCP Tools
55
+
56
+ ### 🤖 Robot Control (8 tools)
57
+
58
+ 1. **`move_robot`** - Move robot to position with memory-guided learning
59
+ 2. **`get_robot_status`** - Get current robot state and sensor data
60
+ 3. **`list_robots`** - Discover available robots
61
+ 4. **`execute_action`** - Execute robot action with strategy
62
+ 5. **`get_sensor_data`** - Read sensor values
63
+ 6. **`set_robot_mode`** - Change robot operation mode
64
+ 7. **`calibrate_robot`** - Run calibration routines
65
+ 8. **`emergency_stop`** - Emergency halt all operations
66
+
67
+ ### 🧠 Memory & Learning (5 tools)
68
+
69
+ 9. **`store_episode`** - Store experience with self-critique
70
+ 10. **`retrieve_memories`** - Semantic search with causal reasoning
71
+ 11. **`consolidate_skills`** - Learn from successful episodes
72
+ 12. **`search_skills`** - Find proven strategies
73
+ 13. **`optimize_memory`** - Database vacuum and reindex
74
+
75
+ ### 🌊 Orchestration (5 tools)
76
+
77
+ 14. **`execute_task`** - Run task with 66 AI agents
78
+ 15. **`execute_swarm`** - Multi-robot parallel coordination
79
+ 16. **`coordinate_robots`** - Strategic task allocation
80
+ 17. **`reason_about_task`** - ReasoningBank for complex decisions
81
+ 18. **`get_orchestration_stats`** - Performance metrics
82
+
83
+ ### 🔬 Advanced (3 tools)
84
+
85
+ 19. **`benchmark_performance`** - Comprehensive performance testing
86
+ 20. **`analyze_memory_patterns`** - Memory usage analytics
87
+ 21. **`synthesize_strategy`** - AI-powered strategy generation
88
+
89
+ ## Architecture
90
+
91
+ ```
92
+ ┌─────────────────────────────────────────┐
93
+ │ MCP Server (stdio) │
94
+ ├─────────────────────────────────────────┤
95
+ │ 🤖 Robot Control │
96
+ │ ├─ Native Bindings (@agentic/core) │
97
+ │ └─ Real-time Communication │
98
+ ├─────────────────────────────────────────┤
99
+ │ 🧠 AgentDB Memory (Hybrid SQL) │
100
+ │ ├─ Reflexion Memory (self-critique) │
101
+ │ ├─ Skill Library (semantic search) │
102
+ │ ├─ Causal Reasoning (cause-effect) │
103
+ │ └─ Performance: 5,725 ops/sec │
104
+ ├─────────────────────────────────────────┤
105
+ │ 🌊 Agentic Flow Orchestration │
106
+ │ ├─ 66 Specialized AI Agents │
107
+ │ ├─ 213 MCP Tools │
108
+ │ ├─ ReasoningBank (learning) │
109
+ │ └─ Multi-Robot Swarm Intelligence │
110
+ └─────────────────────────────────────────┘
111
+ ```
112
+
113
+ ## Performance
114
+
115
+ ### AgentDB Optimization
116
+
117
+ | Operation | CLI Baseline | Hybrid (SQL) | Speedup |
118
+ |-----------|-------------|--------------|---------|
119
+ | Store Episode | 2,300ms | **0.175ms** | **13,168x** |
120
+ | Bulk Store | 2,300ms | **0.008ms** | **271,205x** |
121
+ | Retrieve | 2,000ms | **0.334ms** | **5,988x** |
122
+ | Search Skills | 1,800ms | **0.512ms** | **3,516x** |
123
+
124
+ ### Throughput
125
+
126
+ - **Storage**: 5,725 ops/sec
127
+ - **Bulk Operations**: 117,915 ops/sec
128
+ - **Cache Hit Rate**: 87.3%
129
+ - **Average Latency**: < 1ms
130
+
131
+ ## Usage Examples
132
+
133
+ ### Move Robot with Memory
134
+
135
+ ```javascript
136
+ // Claude can now use this through MCP
137
+ await moveRobot({
138
+ x: 10,
139
+ y: 5,
140
+ z: 0,
141
+ speed: 0.5,
142
+ useMemory: true // Learns from past movements
143
+ });
144
+ ```
145
+
146
+ ### Execute Multi-Robot Task
147
+
148
+ ```javascript
149
+ await executeSwarm({
150
+ robots: ['robot1', 'robot2', 'robot3'],
151
+ tasks: [
152
+ { type: 'patrol', area: 'warehouse-A' },
153
+ { type: 'inspect', area: 'warehouse-B' },
154
+ { type: 'transport', from: 'dock', to: 'storage' }
155
+ ],
156
+ coordination: 'optimal'
157
+ });
158
+ ```
159
+
160
+ ### Learn from Experience
161
+
162
+ ```javascript
163
+ // Store episode with self-critique
164
+ await storeEpisode({
165
+ taskName: 'navigation_obstacle_avoidance',
166
+ success: true,
167
+ confidence: 0.95,
168
+ outcome: 'Successfully navigated around obstacle',
169
+ strategy: 'dynamic_replanning',
170
+ critique: 'Could improve path smoothness'
171
+ });
172
+
173
+ // Retrieve similar situations
174
+ const memories = await retrieveMemories(
175
+ 'obstacle avoidance in narrow corridor',
176
+ { k: 5, synthesizeContext: true }
177
+ );
178
+ ```
179
+
180
+ ### Coordinate Robot Swarm
181
+
182
+ ```javascript
183
+ const plan = await coordinateRobots({
184
+ robots: ['bot1', 'bot2', 'bot3', 'bot4'],
185
+ mission: {
186
+ type: 'warehouse_inventory',
187
+ objectives: [
188
+ 'scan_all_shelves',
189
+ 'detect_misplaced_items',
190
+ 'update_database'
191
+ ],
192
+ deadline: '30 minutes'
193
+ }
194
+ });
195
+
196
+ // Returns optimized task assignments and execution plan
197
+ ```
198
+
199
+ ## Configuration
200
+
201
+ ### Environment Variables
202
+
203
+ ```bash
204
+ # Database path
205
+ export AGENTDB_PATH="./robot-memory.db"
206
+
207
+ # Agentic Flow config
208
+ export AGENTIC_FLOW_AGENTS=66
209
+ export AGENTIC_FLOW_TOOLS=213
210
+
211
+ # Performance tuning
212
+ export MCP_MAX_WORKERS=4
213
+ export MCP_CACHE_SIZE=1000
214
+ ```
215
+
216
+ ### Custom Configuration
217
+
218
+ ```javascript
219
+ // config.json
220
+ {
221
+ "agentdb": {
222
+ "path": "./memory.db",
223
+ "optimization": {
224
+ "walMode": true,
225
+ "cacheSize": 10000,
226
+ "mmapSize": 30000000000
227
+ }
228
+ },
229
+ "agenticFlow": {
230
+ "agents": 66,
231
+ "tools": 213,
232
+ "timeout": 30000
233
+ },
234
+ "swarm": {
235
+ "maxRobots": 10,
236
+ "coordinationStrategy": "optimal"
237
+ }
238
+ }
239
+ ```
240
+
241
+ ## Benchmarking
242
+
243
+ Run comprehensive benchmarks:
244
+
245
+ ```bash
246
+ npm run benchmark:hybrid
247
+ ```
248
+
249
+ Output:
250
+ ```
251
+ 📊 AGENTDB BENCHMARKS (SQL Storage + CLI Search)
252
+
253
+ Store Episode (SQL):
254
+ Iterations: 1,000
255
+ Avg Time: 0.175ms
256
+ Ops/sec: 5,725
257
+ Speedup: 13,168x faster than CLI
258
+
259
+ Bulk Store (SQL Transaction):
260
+ Iterations: 10
261
+ Avg Time: 0.008ms per episode
262
+ Ops/sec: 117,915
263
+ Speedup: 271,205x faster than CLI
264
+ ```
265
+
266
+ ## API Reference
267
+
268
+ See full documentation at [ruv.io/docs](https://ruv.io/docs)
269
+
270
+ ## Requirements
271
+
272
+ - Node.js >= 18.0.0
273
+ - @agentic-robotics/core ^0.1.3
274
+ - agentdb ^1.6.1
275
+ - agentic-flow ^1.10.2
276
+ - better-sqlite3 ^12.4.1
277
+
278
+ ## Related Packages
279
+
280
+ - **[agentic-robotics](https://www.npmjs.com/package/agentic-robotics)** - Complete framework
281
+ - **[@agentic-robotics/core](https://www.npmjs.com/package/@agentic-robotics/core)** - Core bindings
282
+ - **[@agentic-robotics/cli](https://www.npmjs.com/package/@agentic-robotics/cli)** - CLI tools
283
+
284
+ ## Homepage
285
+
286
+ Visit [ruv.io](https://ruv.io) for more information and documentation.
287
+
288
+ ## Contributing
289
+
290
+ Contributions welcome! Visit [github.com/ruvnet/vibecast](https://github.com/ruvnet/vibecast)
291
+
292
+ ## License
293
+
294
+ MIT OR Apache-2.0
295
+
296
+ ## Support
297
+
298
+ - 📚 [Documentation](https://ruv.io/docs)
299
+ - 🐛 [Issue Tracker](https://github.com/ruvnet/vibecast/issues)
300
+ - 💬 [Discussions](https://github.com/ruvnet/vibecast/discussions)
package/package.json ADDED
@@ -0,0 +1,51 @@
1
+ {
2
+ "name": "@agentic-robotics/mcp",
3
+ "version": "0.1.3",
4
+ "description": "Model Context Protocol server for agentic robotics with AgentDB integration",
5
+ "main": "dist/index.js",
6
+ "types": "dist/index.d.ts",
7
+ "bin": {
8
+ "agentic-robotics-mcp": "./dist/cli.js"
9
+ },
10
+ "repository": {
11
+ "type": "git",
12
+ "url": "git+https://github.com/ruvnet/vibecast.git"
13
+ },
14
+ "license": "MIT OR Apache-2.0",
15
+ "keywords": [
16
+ "robotics",
17
+ "mcp",
18
+ "model-context-protocol",
19
+ "agentdb",
20
+ "agentic-flow"
21
+ ],
22
+ "engines": {
23
+ "node": ">= 18"
24
+ },
25
+ "publishConfig": {
26
+ "access": "public"
27
+ },
28
+ "scripts": {
29
+ "build": "tsc",
30
+ "dev": "tsx src/cli.ts",
31
+ "start": "node dist/cli.js",
32
+ "test": "vitest",
33
+ "benchmark": "tsx src/benchmark.ts",
34
+ "benchmark:hybrid": "tsx src/hybrid-benchmark.ts"
35
+ },
36
+ "dependencies": {
37
+ "@agentic-robotics/core": "^0.1.3",
38
+ "@modelcontextprotocol/sdk": "^1.0.0",
39
+ "agentdb": "^1.6.1",
40
+ "agentic-flow": "^1.10.2",
41
+ "better-sqlite3": "^12.4.1"
42
+ },
43
+ "devDependencies": {
44
+ "@types/better-sqlite3": "^7.6.13",
45
+ "@types/node": "^22.0.0",
46
+ "tsx": "^4.19.0",
47
+ "typescript": "^5.7.0",
48
+ "vitest": "^2.1.0"
49
+ },
50
+ "homepage": "https://ruv.io"
51
+ }