방글라데시 콕스바자르를 위한 여행 계획 도구와 날씨 정보를 제공하는 모델 컨텍스트 프로토콜(MCP) 서버입니다. FastMCP로 구축되었으며 uv로 관리됩니다.
Related MCP server: Travel Planner MCP Server
날씨 리소스: 기온 예보 및 상세 날씨 정보
일정 도구: AI 기반 여행 일정 생성
여행 프롬프트: 여행 계획을 위한 사전 구성된 프롬프트
인증 지원: Clerk을 통한 선택적 인증 (환경 변수로 구성 가능)
속도 제한: 내장된 속도 제한 미들웨어
Docker 준비 완료: 프로덕션용 Dockerfile 포함
린팅 및 포맷팅: Ruff + pre-commit 훅 (_docs/lint-formatting.md 참조)
Python 3.13+
uv (패키지 관리자)
Node.js 20+ (MCP Inspector용)
# Install dependencies
uv sync
# Copy environment variables and configure
cp .env.example .env
# (Optional) Install pre-commit git hooks
uv run pre-commit-install모든 명령어는 pyproject.toml에 등록되어 있으며 uv run을 통해 사용할 수 있습니다:
명령어 | 설명 |
| MCP 서버 시작 |
| 개발 모드에서 MCP 서버 시작 (자동 재시작) |
| MCP Inspector UI 실행 (Node.js 20+ 필요) |
| 모든 파일에 대해 pre-commit 훅 실행 (린트 + 포맷) |
| git 저장소에 pre-commit 훅 설치 |
watchdog을 통해 자동 재시작 기능이 포함된 MCP 서버를 시작합니다:
uv run mcp-server-dev
# or
./scripts/run-mcp-server.sh도구, 리소스 및 프롬프트를 테스트하기 위해 대화형 MCP Inspector UI를 실행합니다:
uv run mcp-inspector
# or
./scripts/run-inspector.sh# Run lint + format via pre-commit
uv run lint
# Or run individually
./scripts/lint.sh # ruff check . --fix
./scripts/format.sh # ruff format .전체 구성 세부 정보는 _docs/lint-formatting.md를 참조하세요.
./scripts/test.sh테스트 규칙 및 픽스처는 _docs/testing.md를 참조하세요.
docker build -t mcp-server .
docker run mcp-server서버는 컨테이너 내부에서 uv run mcp-server를 통해 실행됩니다. 전송 방식과 포트는 환경 변수(TRANSPORT_NAME, SERVER_PORT, SERVER_HOST)를 통해 구성할 수 있습니다.
.
├── src/mcp_server/
│ ├── server.py # Main server entry point
│ ├── mcp_instance.py # FastMCP instance & auth config
│ ├── cli.py # CLI command definitions
│ ├── config/
│ │ ├── auth_provider.py # Auth provider factory
│ │ └── custom_routes.py # Custom HTTP routes
│ ├── handlers/ # MCP handler registrations (auto-discovered)
│ │ ├── tools/
│ │ │ ├── auth_additional.py
│ │ │ └── itinerary.py
│ │ ├── resources/
│ │ │ └── weather.py
│ │ └── prompts/
│ │ └── travel_prompts.py
│ ├── models/
│ │ └── itinerary_models.py # Pydantic models & schemas
│ ├── services/
│ │ └── itenerary_service.py # Business logic
│ ├── lib/
│ │ ├── clerk_auth_provider.py # Clerk OAuth provider
│ │ └── httpx_client.py # Async HTTP client wrapper
│ ├── prompt_templates/
│ │ └── travel.py # Prompt text builders
│ └── utils/
│ ├── elicitation.py
│ ├── get_weather_forecast.py
│ ├── helpers.py
│ └── http.py
├── tests/
│ ├── conftest.py
│ ├── fixtures/
│ │ ├── context.py
│ │ └── weather.py
│ ├── unit/
│ │ ├── test_auth_additional_tools.py
│ │ ├── test_auth_provider.py
│ │ ├── test_elicitation.py
│ │ ├── test_helpers.py
│ │ ├── test_itinerary_service_extra.py
│ │ ├── test_itinerary_tool_handler.py
│ │ ├── test_models.py
│ │ ├── test_server.py
│ │ ├── test_travel_prompts.py
│ │ ├── test_travel_prompts_handler.py
│ │ ├── test_weather_forecast.py
│ │ └── test_weather_resource.py
│ └── integration/
│ ├── test_itinerary_tool.py
│ └── test_weather_api.py
├── scripts/
│ ├── run-mcp-server.sh # Dev server with auto-reload
│ ├── run-inspector.sh # MCP Inspector launcher
│ ├── test.sh # Test runner
│ ├── lint.sh # Ruff lint --fix
│ ├── format.sh # Ruff format
│ └── generate-secrets.sh # Secret key generator
├── _docs/ # Documentation & ADRs
│ ├── adr/
│ │ ├── 001-choose-fastmcp.md
│ │ ├── 002-choose-httpx.md
│ │ └── ADR-template.md
│ ├── auth-provider-auth0.md
│ ├── httpx-client.md
│ ├── lint-formatting.md
│ ├── remote-mcp-connect.md
│ └── testing.md
├── .env.example # Environment variables template
├── .pre-commit-config.yaml # Pre-commit hook config
├── Dockerfile # Production Docker image
├── pyproject.toml # Project config & dependencies
├── ruff.toml # Ruff linter/formatter config
├── pytest.ini # Pytest configuration
├── glama.json # Glama registry config
└── LICENSE # MIT License문서 | 설명 |
Ruff 및 pre-commit 구성 | |
테스트 설정, 픽스처 및 규칙 | |
비동기 HTTP 클라이언트 사용법 | |
인증 제공자 통합 | |
원격 MCP 연결 가이드 | |
아키텍처 결정 기록 |
MIT
Full workflow: fetch daily temperatures + generate AI itinerary. Uses the registered MCP prompt 'generate_itinerary' for consistency.
Args: days: Number of days for the trip start_date: Start date (e.g., "2025-01-15", "15 Jan 2025", "today")
Returns: Formatted prompt for AI to generate detailed itinerary
| Name | Required | Description | Default |
|---|---|---|---|
| days | Yes | ||
| start_date | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden. It discloses the tool's workflow and mentions using a registered prompt for consistency, which adds useful context. However, it doesn't cover important behavioral aspects like error handling, rate limits, authentication needs, or what happens if temperature data is unavailable. The description doesn't contradict any annotations since none exist.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured with clear sections: workflow overview, args, and returns. Each sentence adds value, though the 'Full workflow' line could be more concise. The bullet-point format for args and returns is efficient. It's appropriately sized for a 2-parameter tool with a specific workflow.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity (2 parameters, workflow involving external data fetch and AI generation), no annotations, but with an output schema (implied by 'Returns' section), the description is reasonably complete. It explains the purpose, parameters, and output format. The main gap is lack of error handling or edge case guidance, but the output schema reduces the need to fully document return values.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It provides clear semantic meaning for both parameters: 'days' as 'Number of days for the trip' and 'start_date' with format examples. This adds significant value beyond the bare schema, though it doesn't explain constraints like date ranges or day limits. With 0% schema coverage and 2 parameters, this is strong but not perfect compensation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'fetch daily temperatures + generate AI itinerary' and mentions using a registered MCP prompt. It distinguishes from the sibling 'get_activity_suggestions' by focusing on full itinerary generation rather than just suggestions. However, it doesn't specify the exact resource being fetched (e.g., temperatures for what location?), making it slightly less specific than a perfect 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for trip planning with temperature data, but doesn't explicitly state when to use this tool versus alternatives like 'get_activity_suggestions'. It mentions the workflow but lacks clear guidance on prerequisites or exclusions (e.g., whether location data is needed elsewhere).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Suggest activities based on temperature and time of day.
Args: temperature: Temperature in Celsius time_of_day: "morning", "afternoon", or "evening"
Returns: List of suggested activities
| Name | Required | Description | Default |
|---|---|---|---|
| temperature | Yes | ||
| time_of_day | No | afternoon |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It states the tool returns a 'List of suggested activities,' which hints at read-only behavior, but it does not disclose any traits like whether it's safe, if there are rate limits, authentication needs, or how the suggestions are generated. The description is minimal and lacks critical behavioral context for a tool with no annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded, with a clear purpose statement followed by structured sections for 'Args' and 'Returns.' Every sentence earns its place by providing essential information without waste, making it easy to scan and understand quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's low complexity (2 parameters, no annotations, no output schema), the description is somewhat complete but has gaps. It covers the purpose and parameters well, but lacks usage guidelines and behavioral transparency. Without an output schema, it minimally describes returns, but more context on behavior would improve completeness for effective agent use.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description adds significant meaning beyond the input schema, which has 0% description coverage. It explains that 'temperature' is in Celsius and 'time_of_day' can be 'morning', 'afternoon', or 'evening', providing semantic context that the schema lacks. Since there are only 2 parameters and the description compensates well for the low schema coverage, this earns a high score.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Suggest activities based on temperature and time of day.' It specifies the verb ('suggest') and resources ('activities'), and while it doesn't explicitly differentiate from the sibling tool 'cox_ai_itinerary', the purpose is specific enough to understand its function. It's not a tautology since it elaborates beyond just the name.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It mentions the sibling tool 'cox_ai_itinerary' in the context, but the description itself does not indicate any relationship, exclusions, or prerequisites. Usage is implied only by the parameters, with no explicit context or alternatives stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0cox_ai_itineraryget_activity_suggestionsScored across 2 tools
The two tools have clearly distinct purposes: 'cox_ai_itinerary' generates a full multi-day itinerary based on dates and duration, while 'get_activity_suggestions' provides activity recommendations based on specific weather and time conditions. There is no overlap in functionality or ambiguity between them.
The naming is mixed: 'cox_ai_itinerary' uses a descriptive noun phrase with underscores, while 'get_activity_suggestions' follows a verb_noun pattern. Although both are readable, they lack a consistent convention, which could confuse agents expecting a uniform style.
With only two tools, the server feels thin for an itinerary planning domain. It lacks essential operations like updating itineraries, fetching historical data, or managing user preferences, which are typical for such a purpose. This minimal set limits functionality and may require agents to work around gaps.
The tool surface is significantly incomplete for itinerary planning. While it covers itinerary generation and activity suggestions, it misses core CRUD operations (e.g., no way to retrieve, modify, or delete itineraries) and lacks integration with user inputs or preferences. This will likely cause agent failures in real-world scenarios.
Your personal AI travel concierge — flights, hotels, 116M+ POIs, visas, weather & more
Travel tools for AI agents: plan and edit real trips, search stays and tours, import travel videos.
Get current weather for any city and create images from your prompts. Streamline planning, reports…
Travel planner: create, edit, and explore trip itineraries from your Travolp AI assistant.
