Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@context-aware-mcp fetch order ordr_123 details"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A configurable MCP (Model Context Protocol) server with intelligent API routing across microservices. Define your services in JSON, and the server automatically routes natural language queries to the right endpoint.
# Initialize config with a template
npx context-aware-mcp init --template openstore
# Or start with a blank config
npx context-aware-mcp initRelated MCP server: tool-smith
npm install -g context-aware-mcpOr use directly with npx:
npx context-aware-mcp <command>Command | Description |
| Create a new |
| Add a service to config |
| Add an environment URL to a service |
| Set swagger path for a service |
| Remove a service |
| Remove an environment from a service |
| List configured services |
| Validate |
| Compile TypeScript |
| Generate MCP config for Claude Code or VS Code |
All config commands support both --flags mode and interactive mode (run without flags for prompts).
Services are defined in mcp-config.json:
{
"$schema": "./mcp-config.schema.json",
"version": "1.0",
"defaults": { "environment": "qa", "swaggerPath": "/api/docs-json" },
"services": {
"my-service": {
"keywords": ["order", "checkout"],
"idPrefixes": ["ordr_"],
"type": "rest",
"environments": {
"qa": { "url": "https://qa-my-service.example.com" },
"local": { "url": "http://localhost:3001" }
}
}
},
"environmentKeywords": {
"qa": ["from qa", "in qa"],
"local": ["locally", "from local"]
}
}Generate Claude Code or VS Code config:
# Project-level Claude Code config (.mcp.json)
npx context-aware-mcp install
# VS Code config (.vscode/mcp.json)
npx context-aware-mcp install --target vscode
# Both
npx context-aware-mcp install --target allThe server exposes these tools to AI assistants:
smart_api_call — Intelligent routing from natural language queries
fetch_swagger — Fetch and cache OpenAPI specs
call_api — Direct API calls to service endpoints
list_services — List available services
kafka_list_topics / kafka_describe_topic / kafka_read_messages / kafka_consumer_groups — Kafka operations
npm install
npm run build
npm testSee CONTRIBUTING.md for details.
MIT
Direct API call to a specific service endpoint.
Inputs: service (required), endpoint (required), body, method, +3 more.
Fetch OpenAPI spec for a service. Services: os-order
Inputs: service (required), listPaths, clearCache, environment.
List consumer groups and their offsets. Check consumer lag and group status.
Inputs: topic, groupId.
Get detailed metadata about a Kafka topic including partitions, replicas, and current offsets.
Inputs: topic (required).
List all Kafka topics. Optionally filter by pattern (regex).
Inputs: pattern.
Read recent messages from a Kafka topic. Useful for debugging event flows.
Inputs: topic (required), limit, partition, fromBeginning.
List available services and their keywords
Inputs: environment, showKeywords.
Intelligent API routing from natural language queries. Available services: os-order. Environments: local, qa.
Inputs: query (required), body, headers, autoExecute.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 8 tools
The Kafka tools are clearly distinct, but the API tools overlap: smart_api_call and call_api both invoke APIs, and fetch_swagger/list_services are related discovery utilities. Descriptions partially clarify the difference between natural-language routing and direct calls, but an agent could still pick the wrong tool without additional context.
Most tools follow a verb_noun pattern (fetch_swagger, call_api, list_services) and the Kafka group consistently uses kafka_verb_noun. smart_api_call breaks the pattern because it is an adjective-noun phrase rather than a verb-initial name, but the inconsistency is minor.
Eight tools is well-scoped for a server covering both API discovery/invocation and Kafka inspection. Each tool addresses a distinct operational need without redundancy or bloat.
The API side covers service discovery, spec retrieval, and invocation, while the Kafka side covers listing, describing, reading, and consumer-group inspection. Minor gaps such as producing messages or managing offsets exist, but the surface seems complete for its apparent context-gathering purpose.
MCP server for searching Airweave collections with natural language queries.
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
One MCP server for the services your AI needs: memory, files, Microsoft 365, kanban and more.
