gfp-mcp 0.1.0__tar.gz

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.
gfp_mcp-0.1.0/PKG-INFO ADDED
@@ -0,0 +1,360 @@
1
+ Metadata-Version: 2.4
2
+ Name: gfp-mcp
3
+ Version: 0.1.0
4
+ Summary: Model Context Protocol (MCP) server for GDSFactory+ photonic IC design
5
+ Author: GDSFactory+ Team
6
+ License: MIT
7
+ Project-URL: Homepage, https://github.com/doplaydo/gfp-mcp
8
+ Project-URL: Repository, https://github.com/doplaydo/gfp-mcp
9
+ Project-URL: Documentation, https://github.com/doplaydo/gfp-mcp#readme
10
+ Project-URL: Changelog, https://github.com/doplaydo/gfp-mcp/blob/main/CHANGELOG.md
11
+ Project-URL: Issue Tracker, https://github.com/doplaydo/gfp-mcp/issues
12
+ Keywords: mcp,gdsfactory,photonics,ic-design,eda,model-context-protocol,photonic-ic,gds
13
+ Classifier: Development Status :: 4 - Beta
14
+ Classifier: Intended Audience :: Developers
15
+ Classifier: Intended Audience :: Science/Research
16
+ Classifier: Topic :: Scientific/Engineering :: Electronic Design Automation (EDA)
17
+ Classifier: Topic :: Software Development :: Libraries :: Python Modules
18
+ Classifier: License :: OSI Approved :: MIT License
19
+ Classifier: Programming Language :: Python :: 3
20
+ Classifier: Programming Language :: Python :: 3.10
21
+ Classifier: Programming Language :: Python :: 3.11
22
+ Classifier: Programming Language :: Python :: 3.12
23
+ Classifier: Programming Language :: Python :: 3 :: Only
24
+ Requires-Python: >=3.10
25
+ Description-Content-Type: text/markdown
26
+ Requires-Dist: mcp>=1.7.1
27
+ Requires-Dist: httpx>=0.25.0
28
+ Requires-Dist: typing-extensions>=4.0.0; python_version < "3.11"
29
+ Requires-Dist: psutil>=5.9.0
30
+ Provides-Extra: dev
31
+ Requires-Dist: pytest>=7.0.0; extra == "dev"
32
+ Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
33
+ Requires-Dist: pytest-asyncio>=0.21.0; extra == "dev"
34
+ Requires-Dist: ruff>=0.1.0; extra == "dev"
35
+
36
+ # GDSFactory+ MCP Server
37
+
38
+ [![PyPI version](https://img.shields.io/pypi/v/gfp-mcp.svg)](https://pypi.org/project/gfp-mcp/)
39
+ [![Python versions](https://img.shields.io/pypi/pyversions/gfp-mcp.svg)](https://pypi.org/project/gfp-mcp/)
40
+ [![Tests](https://github.com/doplaydo/gfp-mcp/workflows/Tests/badge.svg)](https://github.com/doplaydo/gfp-mcp/actions)
41
+ [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
42
+
43
+ Model Context Protocol (MCP) server for GDSFactory+ that exposes photonic IC design operations as tools for AI assistants like Claude Code and Claude Desktop.
44
+
45
+ ## Overview
46
+
47
+ This project implements a standalone MCP server that bridges AI assistants with GDSFactory+ photonic IC design capabilities. The server acts as a lightweight proxy that exposes GDSFactory+ operations through standardized MCP tools while maintaining zero modifications to the existing FastAPI backend.
48
+
49
+ ## Features
50
+
51
+ ### Phase 1: Core Building Tools (Complete)
52
+
53
+ - **build_cell** - Build a single GDS cell by name
54
+ - **build_cells** - Build multiple GDS cells in batch
55
+ - **list_cells** - List all available photonic components
56
+ - **get_cell_info** - Get detailed component metadata
57
+ - **download_gds** - Download built GDS files
58
+ - **list_projects** - List all running GDSFactory+ server instances
59
+ - **get_project_info** - Get detailed information about a specific project
60
+
61
+ ### Multi-Project Support
62
+
63
+ The MCP server integrates with the GDSFactory+ server registry to support working with multiple projects simultaneously. The registry is stored at `~/.gdsfactory/server-registry.json` and is automatically managed by GDSFactory+ servers.
64
+
65
+ #### How It Works
66
+
67
+ 1. When you start a GDSFactory+ server with `gfp serve`, it registers itself in the shared registry
68
+ 2. The MCP server reads from this registry to discover available projects
69
+ 3. Tools accept an optional `project` parameter to route requests to specific servers
70
+ 4. The MCP server automatically resolves project names to the correct port
71
+
72
+ #### Example Usage
73
+
74
+ ```
75
+ User: "List all running GDSFactory+ projects"
76
+ Claude: [Uses list_projects tool to show all registered servers]
77
+
78
+ User: "Build the mzi component in the my_photonics_project"
79
+ Claude: [Uses build_cell tool with project="my_photonics_project"]
80
+ ```
81
+
82
+ **Note**: The MCP has read-only access to the registry. Only GDSFactory+ servers can register/unregister themselves.
83
+
84
+ ### Architecture
85
+
86
+ ```
87
+ AI Assistant (Claude) <-> MCP Server (STDIO) <-> HTTP Client <-> FastAPI Server
88
+ ```
89
+
90
+ **Key Benefits:**
91
+ - Universal MCP client compatibility via STDIO transport
92
+ - Zero modifications to existing FastAPI server
93
+ - Clean separation of concerns
94
+ - No database conflicts (only FastAPI touches SQLite)
95
+ - Scalable architecture ready for 20+ tools
96
+
97
+ ## Installation
98
+
99
+ ### From PyPI (Recommended)
100
+
101
+ Install the package from PyPI:
102
+
103
+ ```bash
104
+ pip install gfp-mcp
105
+ ```
106
+
107
+ Or with uv:
108
+
109
+ ```bash
110
+ uv pip install gfp-mcp
111
+ ```
112
+
113
+ ### From Source (Development)
114
+
115
+ For development or if you want the latest unreleased changes:
116
+
117
+ ```bash
118
+ git clone https://github.com/doplaydo/gfp-mcp.git
119
+ cd gfp-mcp
120
+ pip install -e ".[dev]"
121
+ ```
122
+
123
+ ### Prerequisites
124
+
125
+ - Python 3.10 or higher
126
+ - GDSFactory+ with FastAPI server running
127
+
128
+ ### Verify Installation
129
+
130
+ ```bash
131
+ gfp-mcp-serve --help
132
+ ```
133
+
134
+ ## Quick Start
135
+
136
+ ### 1. Start the FastAPI Server
137
+
138
+ In one terminal:
139
+
140
+ ```bash
141
+ gfp serve --port 8787
142
+ ```
143
+
144
+ ### 2. Configure Your AI Assistant
145
+
146
+ #### Claude Code
147
+
148
+ Add to `.claude/settings.json`:
149
+
150
+ ```json
151
+ {
152
+ "mcpServers": {
153
+ "gdsfactoryplus": {
154
+ "command": "gfp-mcp-serve",
155
+ "args": [],
156
+ "env": {
157
+ "GFP_API_URL": "http://localhost:8787"
158
+ }
159
+ }
160
+ }
161
+ }
162
+ ```
163
+
164
+ Or use the command line:
165
+
166
+ ```bash
167
+ claude mcp add gdsfactoryplus -- gfp-mcp-serve
168
+ ```
169
+
170
+ #### Claude Desktop
171
+
172
+ Add to `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS):
173
+
174
+ ```json
175
+ {
176
+ "mcpServers": {
177
+ "gdsfactoryplus": {
178
+ "command": "gfp-mcp-serve",
179
+ "args": []
180
+ }
181
+ }
182
+ }
183
+ ```
184
+
185
+ For Windows: `%APPDATA%\Claude\claude_desktop_config.json`
186
+
187
+ For Linux: `~/.config/Claude/claude_desktop_config.json`
188
+
189
+ ### 3. Use the Tools
190
+
191
+ Ask your AI assistant to:
192
+
193
+ - "List all available photonic components"
194
+ - "Build the mzi component"
195
+ - "Show me details about the coupler component"
196
+ - "Build multiple components: mzi, coupler, and bend_euler"
197
+
198
+ ## Environment Variables
199
+
200
+ Configure the MCP server using environment variables:
201
+
202
+ ```bash
203
+ # FastAPI server URL (default: http://localhost:8787)
204
+ export GFP_API_URL="http://localhost:8787"
205
+
206
+ # Request timeout in seconds (default: 300)
207
+ export GFP_MCP_TIMEOUT=300
208
+
209
+ # Enable debug logging (default: false)
210
+ export GFP_MCP_DEBUG=true
211
+ ```
212
+
213
+ ## Project Structure
214
+
215
+ ```
216
+ gfp-mcp/
217
+ ├── mcp_standalone/ # MCP server implementation
218
+ │ ├── __init__.py # Package exports
219
+ │ ├── config.py # Configuration management
220
+ │ ├── client.py # HTTP client for FastAPI
221
+ │ ├── registry.py # Server registry (multi-project support)
222
+ │ ├── tools.py # MCP tool definitions
223
+ │ ├── mappings.py # Tool → Endpoint mappings
224
+ │ ├── server.py # MCP server core
225
+ │ └── README.md # Detailed documentation
226
+ ├── tests/ # Test suite
227
+ │ ├── test_mcp_tools.py
228
+ │ ├── test_mcp_mappings.py
229
+ │ ├── test_mcp_integration.py
230
+ │ └── test_registry.py
231
+ ├── mcp_serve.py # CLI entry point
232
+ ├── MCP_QUICKSTART.md # Quick start guide
233
+ ├── MCP_IMPLEMENTATION_STATUS.md # Implementation details
234
+ └── README.md # This file
235
+ ```
236
+
237
+ ## Testing
238
+
239
+ Run the test suite:
240
+
241
+ ```bash
242
+ # Run all MCP tests
243
+ pytest tests/test_mcp_*.py -v
244
+
245
+ # Run with coverage
246
+ pytest tests/test_mcp_*.py --cov=mcp_standalone --cov-report=term-missing
247
+
248
+ # Run specific test file
249
+ pytest tests/test_mcp_tools.py -v
250
+ ```
251
+
252
+ All tests are passing (46/46 tests):
253
+ - Tool definitions: 15 tests
254
+ - Endpoint mappings: 13 tests
255
+ - Integration & client: 9 tests
256
+ - Registry integration: 9 tests
257
+
258
+ ## Example Workflows
259
+
260
+ ### Build and Verify a Component
261
+
262
+ ```
263
+ User: "List all available photonic components"
264
+ Claude: [Uses list_cells tool]
265
+
266
+ User: "Build the mzi component"
267
+ Claude: [Uses build_cell tool with name="mzi"]
268
+
269
+ User: "Show me the details of the mzi component"
270
+ Claude: [Uses get_cell_info tool with name="mzi"]
271
+ ```
272
+
273
+ ### Batch Build Multiple Components
274
+
275
+ ```
276
+ User: "Build the following components: mzi, coupler, and bend_euler"
277
+ Claude: [Uses build_cells tool with names=["mzi", "coupler", "bend_euler"]]
278
+ ```
279
+
280
+ ## Troubleshooting
281
+
282
+ ### Error: "Connection refused to localhost:8787"
283
+
284
+ **Cause**: FastAPI server is not running
285
+
286
+ **Solution**: Start the FastAPI server:
287
+
288
+ ```bash
289
+ gfp serve --port 8787
290
+ ```
291
+
292
+ ### Error: "MCP server not responding"
293
+
294
+ **Cause**: STDIO transport issue or MCP client misconfiguration
295
+
296
+ **Solution**:
297
+ 1. Check Claude Code/Desktop logs with `claude --debug`
298
+ 2. Restart the MCP server
299
+ 3. Verify the configuration in settings.json
300
+
301
+ ### Error: "Tool execution timeout"
302
+
303
+ **Cause**: Long-running operation exceeded timeout
304
+
305
+ **Solution**: Increase the timeout:
306
+
307
+ ```bash
308
+ export GFP_MCP_TIMEOUT=600 # 10 minutes
309
+ gfp mcp-serve
310
+ ```
311
+
312
+ ## Roadmap
313
+
314
+ ### Future Phases
315
+
316
+ - **Phase 2**: Verification tools (DRC, LVS)
317
+ - **Phase 3**: SPICE workflow tools
318
+ - **Phase 4**: Simulation & advanced tools
319
+ - **Phase 5**: Comprehensive testing & documentation
320
+
321
+ ## Documentation
322
+
323
+ - [Quick Start Guide](MCP_QUICKSTART.md) - Step-by-step setup instructions
324
+ - [Implementation Status](MCP_IMPLEMENTATION_STATUS.md) - Detailed implementation notes
325
+ - [MCP Standalone README](mcp_standalone/README.md) - Architecture details
326
+ - [Contributing Guide](CONTRIBUTING.md) - Development and release guidelines
327
+ - [Changelog](CHANGELOG.md) - Version history and release notes
328
+
329
+ ## Contributing
330
+
331
+ Contributions are welcome! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for detailed guidelines on:
332
+
333
+ - Development setup
334
+ - Running tests and code quality checks
335
+ - Making changes and submitting PRs
336
+ - Release process (for maintainers)
337
+
338
+ Quick checklist:
339
+
340
+ 1. All tests pass: `pytest tests/test_mcp_*.py -v`
341
+ 2. Code quality: `ruff check . && ruff format --check .`
342
+ 3. Documentation is updated
343
+ 4. Changes are backward compatible
344
+
345
+ ## License
346
+
347
+ See the main GDSFactory+ project for license information.
348
+
349
+ ## Support
350
+
351
+ For issues or questions:
352
+
353
+ 1. Check the [Quick Start Guide](MCP_QUICKSTART.md) troubleshooting section
354
+ 2. Review the [Implementation Status](MCP_IMPLEMENTATION_STATUS.md) document
355
+ 3. Enable debug mode with `GFP_MCP_DEBUG=true` and check server logs
356
+ 4. Open an issue on GitHub
357
+
358
+ ## Acknowledgments
359
+
360
+ Built on the Model Context Protocol by Anthropic, enabling seamless AI assistant integration with photonic IC design workflows.
@@ -0,0 +1,325 @@
1
+ # GDSFactory+ MCP Server
2
+
3
+ [![PyPI version](https://img.shields.io/pypi/v/gfp-mcp.svg)](https://pypi.org/project/gfp-mcp/)
4
+ [![Python versions](https://img.shields.io/pypi/pyversions/gfp-mcp.svg)](https://pypi.org/project/gfp-mcp/)
5
+ [![Tests](https://github.com/doplaydo/gfp-mcp/workflows/Tests/badge.svg)](https://github.com/doplaydo/gfp-mcp/actions)
6
+ [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
7
+
8
+ Model Context Protocol (MCP) server for GDSFactory+ that exposes photonic IC design operations as tools for AI assistants like Claude Code and Claude Desktop.
9
+
10
+ ## Overview
11
+
12
+ This project implements a standalone MCP server that bridges AI assistants with GDSFactory+ photonic IC design capabilities. The server acts as a lightweight proxy that exposes GDSFactory+ operations through standardized MCP tools while maintaining zero modifications to the existing FastAPI backend.
13
+
14
+ ## Features
15
+
16
+ ### Phase 1: Core Building Tools (Complete)
17
+
18
+ - **build_cell** - Build a single GDS cell by name
19
+ - **build_cells** - Build multiple GDS cells in batch
20
+ - **list_cells** - List all available photonic components
21
+ - **get_cell_info** - Get detailed component metadata
22
+ - **download_gds** - Download built GDS files
23
+ - **list_projects** - List all running GDSFactory+ server instances
24
+ - **get_project_info** - Get detailed information about a specific project
25
+
26
+ ### Multi-Project Support
27
+
28
+ The MCP server integrates with the GDSFactory+ server registry to support working with multiple projects simultaneously. The registry is stored at `~/.gdsfactory/server-registry.json` and is automatically managed by GDSFactory+ servers.
29
+
30
+ #### How It Works
31
+
32
+ 1. When you start a GDSFactory+ server with `gfp serve`, it registers itself in the shared registry
33
+ 2. The MCP server reads from this registry to discover available projects
34
+ 3. Tools accept an optional `project` parameter to route requests to specific servers
35
+ 4. The MCP server automatically resolves project names to the correct port
36
+
37
+ #### Example Usage
38
+
39
+ ```
40
+ User: "List all running GDSFactory+ projects"
41
+ Claude: [Uses list_projects tool to show all registered servers]
42
+
43
+ User: "Build the mzi component in the my_photonics_project"
44
+ Claude: [Uses build_cell tool with project="my_photonics_project"]
45
+ ```
46
+
47
+ **Note**: The MCP has read-only access to the registry. Only GDSFactory+ servers can register/unregister themselves.
48
+
49
+ ### Architecture
50
+
51
+ ```
52
+ AI Assistant (Claude) <-> MCP Server (STDIO) <-> HTTP Client <-> FastAPI Server
53
+ ```
54
+
55
+ **Key Benefits:**
56
+ - Universal MCP client compatibility via STDIO transport
57
+ - Zero modifications to existing FastAPI server
58
+ - Clean separation of concerns
59
+ - No database conflicts (only FastAPI touches SQLite)
60
+ - Scalable architecture ready for 20+ tools
61
+
62
+ ## Installation
63
+
64
+ ### From PyPI (Recommended)
65
+
66
+ Install the package from PyPI:
67
+
68
+ ```bash
69
+ pip install gfp-mcp
70
+ ```
71
+
72
+ Or with uv:
73
+
74
+ ```bash
75
+ uv pip install gfp-mcp
76
+ ```
77
+
78
+ ### From Source (Development)
79
+
80
+ For development or if you want the latest unreleased changes:
81
+
82
+ ```bash
83
+ git clone https://github.com/doplaydo/gfp-mcp.git
84
+ cd gfp-mcp
85
+ pip install -e ".[dev]"
86
+ ```
87
+
88
+ ### Prerequisites
89
+
90
+ - Python 3.10 or higher
91
+ - GDSFactory+ with FastAPI server running
92
+
93
+ ### Verify Installation
94
+
95
+ ```bash
96
+ gfp-mcp-serve --help
97
+ ```
98
+
99
+ ## Quick Start
100
+
101
+ ### 1. Start the FastAPI Server
102
+
103
+ In one terminal:
104
+
105
+ ```bash
106
+ gfp serve --port 8787
107
+ ```
108
+
109
+ ### 2. Configure Your AI Assistant
110
+
111
+ #### Claude Code
112
+
113
+ Add to `.claude/settings.json`:
114
+
115
+ ```json
116
+ {
117
+ "mcpServers": {
118
+ "gdsfactoryplus": {
119
+ "command": "gfp-mcp-serve",
120
+ "args": [],
121
+ "env": {
122
+ "GFP_API_URL": "http://localhost:8787"
123
+ }
124
+ }
125
+ }
126
+ }
127
+ ```
128
+
129
+ Or use the command line:
130
+
131
+ ```bash
132
+ claude mcp add gdsfactoryplus -- gfp-mcp-serve
133
+ ```
134
+
135
+ #### Claude Desktop
136
+
137
+ Add to `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS):
138
+
139
+ ```json
140
+ {
141
+ "mcpServers": {
142
+ "gdsfactoryplus": {
143
+ "command": "gfp-mcp-serve",
144
+ "args": []
145
+ }
146
+ }
147
+ }
148
+ ```
149
+
150
+ For Windows: `%APPDATA%\Claude\claude_desktop_config.json`
151
+
152
+ For Linux: `~/.config/Claude/claude_desktop_config.json`
153
+
154
+ ### 3. Use the Tools
155
+
156
+ Ask your AI assistant to:
157
+
158
+ - "List all available photonic components"
159
+ - "Build the mzi component"
160
+ - "Show me details about the coupler component"
161
+ - "Build multiple components: mzi, coupler, and bend_euler"
162
+
163
+ ## Environment Variables
164
+
165
+ Configure the MCP server using environment variables:
166
+
167
+ ```bash
168
+ # FastAPI server URL (default: http://localhost:8787)
169
+ export GFP_API_URL="http://localhost:8787"
170
+
171
+ # Request timeout in seconds (default: 300)
172
+ export GFP_MCP_TIMEOUT=300
173
+
174
+ # Enable debug logging (default: false)
175
+ export GFP_MCP_DEBUG=true
176
+ ```
177
+
178
+ ## Project Structure
179
+
180
+ ```
181
+ gfp-mcp/
182
+ ├── mcp_standalone/ # MCP server implementation
183
+ │ ├── __init__.py # Package exports
184
+ │ ├── config.py # Configuration management
185
+ │ ├── client.py # HTTP client for FastAPI
186
+ │ ├── registry.py # Server registry (multi-project support)
187
+ │ ├── tools.py # MCP tool definitions
188
+ │ ├── mappings.py # Tool → Endpoint mappings
189
+ │ ├── server.py # MCP server core
190
+ │ └── README.md # Detailed documentation
191
+ ├── tests/ # Test suite
192
+ │ ├── test_mcp_tools.py
193
+ │ ├── test_mcp_mappings.py
194
+ │ ├── test_mcp_integration.py
195
+ │ └── test_registry.py
196
+ ├── mcp_serve.py # CLI entry point
197
+ ├── MCP_QUICKSTART.md # Quick start guide
198
+ ├── MCP_IMPLEMENTATION_STATUS.md # Implementation details
199
+ └── README.md # This file
200
+ ```
201
+
202
+ ## Testing
203
+
204
+ Run the test suite:
205
+
206
+ ```bash
207
+ # Run all MCP tests
208
+ pytest tests/test_mcp_*.py -v
209
+
210
+ # Run with coverage
211
+ pytest tests/test_mcp_*.py --cov=mcp_standalone --cov-report=term-missing
212
+
213
+ # Run specific test file
214
+ pytest tests/test_mcp_tools.py -v
215
+ ```
216
+
217
+ All tests are passing (46/46 tests):
218
+ - Tool definitions: 15 tests
219
+ - Endpoint mappings: 13 tests
220
+ - Integration & client: 9 tests
221
+ - Registry integration: 9 tests
222
+
223
+ ## Example Workflows
224
+
225
+ ### Build and Verify a Component
226
+
227
+ ```
228
+ User: "List all available photonic components"
229
+ Claude: [Uses list_cells tool]
230
+
231
+ User: "Build the mzi component"
232
+ Claude: [Uses build_cell tool with name="mzi"]
233
+
234
+ User: "Show me the details of the mzi component"
235
+ Claude: [Uses get_cell_info tool with name="mzi"]
236
+ ```
237
+
238
+ ### Batch Build Multiple Components
239
+
240
+ ```
241
+ User: "Build the following components: mzi, coupler, and bend_euler"
242
+ Claude: [Uses build_cells tool with names=["mzi", "coupler", "bend_euler"]]
243
+ ```
244
+
245
+ ## Troubleshooting
246
+
247
+ ### Error: "Connection refused to localhost:8787"
248
+
249
+ **Cause**: FastAPI server is not running
250
+
251
+ **Solution**: Start the FastAPI server:
252
+
253
+ ```bash
254
+ gfp serve --port 8787
255
+ ```
256
+
257
+ ### Error: "MCP server not responding"
258
+
259
+ **Cause**: STDIO transport issue or MCP client misconfiguration
260
+
261
+ **Solution**:
262
+ 1. Check Claude Code/Desktop logs with `claude --debug`
263
+ 2. Restart the MCP server
264
+ 3. Verify the configuration in settings.json
265
+
266
+ ### Error: "Tool execution timeout"
267
+
268
+ **Cause**: Long-running operation exceeded timeout
269
+
270
+ **Solution**: Increase the timeout:
271
+
272
+ ```bash
273
+ export GFP_MCP_TIMEOUT=600 # 10 minutes
274
+ gfp mcp-serve
275
+ ```
276
+
277
+ ## Roadmap
278
+
279
+ ### Future Phases
280
+
281
+ - **Phase 2**: Verification tools (DRC, LVS)
282
+ - **Phase 3**: SPICE workflow tools
283
+ - **Phase 4**: Simulation & advanced tools
284
+ - **Phase 5**: Comprehensive testing & documentation
285
+
286
+ ## Documentation
287
+
288
+ - [Quick Start Guide](MCP_QUICKSTART.md) - Step-by-step setup instructions
289
+ - [Implementation Status](MCP_IMPLEMENTATION_STATUS.md) - Detailed implementation notes
290
+ - [MCP Standalone README](mcp_standalone/README.md) - Architecture details
291
+ - [Contributing Guide](CONTRIBUTING.md) - Development and release guidelines
292
+ - [Changelog](CHANGELOG.md) - Version history and release notes
293
+
294
+ ## Contributing
295
+
296
+ Contributions are welcome! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for detailed guidelines on:
297
+
298
+ - Development setup
299
+ - Running tests and code quality checks
300
+ - Making changes and submitting PRs
301
+ - Release process (for maintainers)
302
+
303
+ Quick checklist:
304
+
305
+ 1. All tests pass: `pytest tests/test_mcp_*.py -v`
306
+ 2. Code quality: `ruff check . && ruff format --check .`
307
+ 3. Documentation is updated
308
+ 4. Changes are backward compatible
309
+
310
+ ## License
311
+
312
+ See the main GDSFactory+ project for license information.
313
+
314
+ ## Support
315
+
316
+ For issues or questions:
317
+
318
+ 1. Check the [Quick Start Guide](MCP_QUICKSTART.md) troubleshooting section
319
+ 2. Review the [Implementation Status](MCP_IMPLEMENTATION_STATUS.md) document
320
+ 3. Enable debug mode with `GFP_MCP_DEBUG=true` and check server logs
321
+ 4. Open an issue on GitHub
322
+
323
+ ## Acknowledgments
324
+
325
+ Built on the Model Context Protocol by Anthropic, enabling seamless AI assistant integration with photonic IC design workflows.