Defect Inspection Tools MCP Server

A TypeScript-based server for handling defect inspection, analysis, and reporting with AI/ML capabilities for quality control.

Features

• Defect detection and analysis
• Inspection management (SIRE, CDI, PSC)
• Audit and compliance tracking
• Fleet defect monitoring
• Export and reporting capabilities
• Typesense integration for search
• MongoDB integration for data storage

Prerequisites

• Node.js (v18 or higher)
• MongoDB
• Typesense
• OpenAI API key
• LlamaParse API key
• Perplexity API key

Configuration

You can configure the server in three ways:

1. Environment-Specific Configuration Files

The server now supports environment-specific configuration files. Create a file named .env.{environment} where {environment} matches your NODE_ENV value:

.env.development  # Used when NODE_ENV=development (default)
.env.production   # Used when NODE_ENV=production
.env.staging      # Used when NODE_ENV=staging
.env              # Fallback if no environment-specific file is found

To use a specific environment:

NODE_ENV=production node dist/index.js

See the env.production.example file for a template of production environment variables.

2. Environment Variables (.env file)

Create a .env file in the root directory:

MONGO_URI=mongodb://localhost:27017
DB_NAME=mcp_defect_inspection
TYPESENSE_HOST=localhost
TYPESENSE_PORT=8108
TYPESENSE_PROTOCOL=http
TYPESENSE_API_KEY=your_typesense_api_key
OPENAI_API_KEY=your_openai_api_key
LLAMA_API_KEY=your_llama_api_key
VENDOR_MODEL=gpt-4
S3_API_TOKEN=your_s3_api_token
S3_GENERATE_HTML_URL=your_s3_generate_html_url
LLAMA_PARSE_URL=your_llama_parse_url
PERPLEXITY_API_KEY=your_perplexity_api_key
GOOGLE_CLIENT_ID=your_google_client_id
GOOGLE_CLIENT_SECRET=your_google_client_secret
GOOGLE_REDIRECT_URI=your_google_redirect_uri
GOOGLE_API_KEY=your_google_api_key
GOOGLE_SEARCH_ENGINE_ID=your_google_search_engine_id
COHERE_API_KEY=your_cohere_api_key
COMPANY_NAME=your_company_name
# API Configuration
SNAPSHOT_URL=https://dev-api.siya.com/v1.0/vessel-info/qna-snapshot
JWT_TOKEN=your_jwt_token_here

3. Command Line Arguments

Command line arguments will override values from the .env file:

node dist/index.js \
  --mongo-uri mongodb://localhost:27017 \
  --db-name mcp_defect_inspection \
  --typesense-host localhost \
  --typesense-port 8108 \
  --typesense-protocol http \
  --typesense-api-key your_api_key \
  --openai-api-key your_openai_key \
  --llama-api-key your_llama_key \
  --vendor-model gpt-4 \
  --s3-api-token your_s3_token \
  --perplexity-api-key your_perplexity_key \
  --company-name your_company_name

Available Command Line Arguments

| Argument | Environment Variable | Default | Description | |----------|---------------------|---------|-------------| | --mongo-uri | MONGO_URI | mongodb://localhost:27017 | MongoDB connection URI | | --db-name | DB_NAME | mcp_defect_inspection | MongoDB database name | | --typesense-host | TYPESENSE_HOST | localhost | Typesense server host | | --typesense-port | TYPESENSE_PORT | 8108 | Typesense server port | | --typesense-protocol | TYPESENSE_PROTOCOL | http | Typesense protocol (http/https) | | --typesense-api-key | TYPESENSE_API_KEY | (empty) | Typesense API key | | --openai-api-key | OPENAI_API_KEY | (empty) | OpenAI API key | | --llama-api-key | LLAMA_API_KEY | (empty) | LlamaParse API key | | --vendor-model | VENDOR_MODEL | gpt-4 | Default LLM model | | --s3-api-token | S3_API_TOKEN | (empty) | S3 API token | | --s3-generate-html-url | S3_GENERATE_HTML_URL | (empty) | S3 HTML generation URL | | --llama-parse-url | LLAMA_PARSE_URL | (empty) | LlamaParse service URL | | --perplexity-api-key | PERPLEXITY_API_KEY | (empty) | Perplexity API key | | --google-client-id | GOOGLE_CLIENT_ID | (empty) | Google OAuth client ID | | --google-client-secret | GOOGLE_CLIENT_SECRET | (empty) | Google OAuth client secret | | --google-redirect-uri | GOOGLE_REDIRECT_URI | (empty) | Google OAuth redirect URI | | --google-api-key | GOOGLE_API_KEY | (empty) | Google API key | | --google-search-engine-id | GOOGLE_SEARCH_ENGINE_ID | (empty) | Google Custom Search Engine ID | | --cohere-api-key | COHERE_API_KEY | (empty) | Cohere API key | | --company-name | COMPANY_NAME | (required) | Company name (mandatory) | | --snapshot-url | SNAPSHOT_URL | https://dev-api.siya.com/v1.0/vessel-info/qna-snapshot | API endpoint for vessel QnA snapshots | | --jwt-token | JWT_TOKEN | (default dev token) | JWT token for API authentication |

Installation

1. Clone the repository:

git clone https://github.com/syia-ai/defect_inspection_tools_mcp_server.git
cd defect_inspection_tools_mcp_server

2. Install dependencies:

npm install

3. Build the project:

npm run build

4. Start the server:

npm start

Development

For development with hot reload:

npm run dev

Testing

Test the server with MCP Inspector:

npm test

MCP Tools and Resources

Defect Management Tools

• universal_defect_search: Comprehensive defect search with multiple filters
• get_all_vessel_defects_records: Get all defect records for a vessel
• get_summary_of_defects: Get summary statistics of defects
• export_fleet_defects: Export defect data for fleet analysis

Inspection Management Tools

• get_tmsa_summary: Get TMSA (Tanker Management Self Assessment) summary
• get_sire_reports_from_ocimf: Retrieve SIRE reports from OCIMF
• fetch_fleet_historical_sire_observations: Get historical SIRE observations
• get_cdi_reports_from_ocimf: Retrieve CDI reports from OCIMF
• get_psc_inspection_defects: Get Port State Control inspection defects

Audit & Compliance Tools

• get_vir_status_overview: Get Vessel Inspection Report status overview
• get_internal_audit_summary: Get internal audit summary
• get_fleet_sire_cdi_inspection_status: Get fleet SIRE/CDI inspection status
• get_fleet_vir_status_overview: Get fleet VIR status overview
• get_fleet_last_internal_audit_summary: Get fleet last internal audit summary

Project Structure

src/
├── tools/
│   ├── handlers/
│   │   ├── defectTools.ts      # Defect management and analysis
│   │   ├── inspectionTools.ts  # Inspection management (SIRE, CDI, PSC)
│   │   ├── auditTools.ts       # Audit and compliance tools
│   │   ├── exportTools.ts      # Data export functionality
│   │   └── universalTools.ts   # Universal search and utility tools
│   ├── schema.ts          # Tool schema definitions
│   └── index.ts          # Main tool handler
├── config/
│   └── index.ts          # Configuration management
├── types/
│   └── index.ts          # TypeScript type definitions
├── prompts/
│   └── index.ts          # Prompt management
├── resources/
│   └── index.ts          # Resource management
└── index.ts              # Main application entry point

Key Features

Defect Detection & Analysis

• Comprehensive defect search and filtering
• Risk categorization and priority assessment
• Defect tracking and status management
• Historical defect analysis

Inspection Management

• SIRE (Ship Inspection Report) integration
• CDI (Condition Assessment Program) reports
• Port State Control (PSC) inspection tracking
• TMSA (Tanker Management Self Assessment) support

Audit & Compliance

• Vessel Inspection Report (VIR) management
• Internal audit tracking and reporting
• Fleet-wide compliance monitoring
• Regulatory requirement tracking

Vessel Data Integration

• Link defects and inspections to vessels by IMO
• Fleet-wide defect monitoring
• Historical inspection data analysis
• Export capabilities for reporting

Search Integration

• Full-text search using Typesense
• Vector search for semantic matching
• Filtering and pagination support

Error Handling

The application implements comprehensive error handling:

• Input validation
• Database operation error handling
• API integration error handling
• Proper error logging
• User-friendly error messages

Logging

Logging is implemented using Winston that:

• Logs to console with timestamps
• Includes different log levels (debug, info, error)
• Provides detailed error information
• Supports structured logging

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.