mcp-airflow-api 0.0.0__tar.gz

mcp_airflow_api-0.0.0/LICENSE

@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2015 Anthony Lapenna
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

mcp_airflow_api-0.0.0/MANIFEST.in

@@ -0,0 +1 @@
+include src/mcp_airflow_api/prompt_template.md

mcp_airflow_api-0.0.0/PKG-INFO

@@ -0,0 +1,169 @@
+Metadata-Version: 2.4
+Name: mcp-airflow-api
+Version: 0.0.0
+Summary: Model Context Protocol (MCP) server for Apache Airflow API integration. Provides comprehensive tools for managing Airflow clusters including service operations, configuration management, status monitoring, and request tracking.
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: fastapi>=0.116.1
+Requires-Dist: requests>=2.32.4
+Requires-Dist: uvicorn>=0.35.0
+Requires-Dist: mcp>=1.12.3
+Dynamic: license-file
+
+Model Context Protocol (MCP) server for Apache Airflow API integration.  
+This project provides natural language MCP tools for essential Airflow cluster operations.
+
+[![Deploy to PyPI with tag](https://github.com/call518/MCP-Airflow-API/actions/workflows/pypi-publish.yml/badge.svg)](https://github.com/call518/MCP-Airflow-API/actions/workflows/pypi-publish.yml)
+
+[![smithery badge](https://smithery.ai/badge/@call518/mcp-airflow-api)](https://smithery.ai/server/@call518/mcp-airflow-api)
+
+---
+
+
+# MCP-Airflow-API
+
+**Tested and supported Airflow version: 2.10.2 (API Version: v1)**
+
+## Features
+
+- List all DAGs in the Airflow cluster
+- Monitor running/failed DAG runs
+- Trigger DAG runs on demand
+- Minimal, LLM-friendly output for all tools
+- Easy integration with MCP Inspector, OpenWebUI, Smithery, etc.
+
+---
+
+## Available MCP Tools
+
+### DAG Management
+
+- `list_dags`  
+	Returns all DAGs registered in the Airflow cluster.  
+	Output: `dag_id`, `dag_display_name`, `is_active`, `is_paused`, `owners`, `tags`
+
+- `running_dags`  
+	Returns all currently running DAG runs.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `failed_dags`  
+	Returns all recently failed DAG runs.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `trigger_dag(dag_id)`  
+	Immediately triggers the specified DAG.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `pause_dag(dag_id)`  
+	Pauses the specified DAG (prevents scheduling new runs).  
+	Output: `dag_id`, `is_paused`
+
+- `unpause_dag(dag_id)`  
+	Unpauses the specified DAG (allows scheduling new runs).  
+	Output: `dag_id`, `is_paused`
+
+---
+
+## Prompt Template
+
+The package exposes a tool `get_prompt_template` that returns either the entire template, a specific section, or just the headings. Three MCP prompts (`prompt_template_full`, `prompt_template_headings`, `prompt_template_section`) are also registered for discovery.
+
+### MCP Prompts
+
+For easier discoverability in MCP clients (so `prompts/list` is not empty), the server now registers three prompts:
+
+• `prompt_template_full` – returns the full canonical template  
+• `prompt_template_headings` – returns only the section headings  
+• `prompt_template_section` – takes a `section` argument (number or keyword) and returns that section
+
+You can still use the `get_prompt_template` tool for programmatic access or when you prefer tool invocation over prompt retrieval.
+
+Single canonical English prompt template guides safe and efficient tool selection.
+
+Files:
+• Packaged: `src/mcp_airflow_api/prompt_template.md` (distributed with PyPI)  
+• (Optional workspace root copy `PROMPT_TEMPLATE.md` may exist for editing; packaged copy is the one loaded at runtime.)
+
+Retrieve dynamically via MCP tool:
+• `get_prompt_template()` – full template  
+• `get_prompt_template("tool map")` – only the tool mapping section  
+• `get_prompt_template("3")` – section 3 (tool map)  
+• `get_prompt_template(mode="headings")` – list all section headings
+
+Policy: Only English is stored; LLM는 사용자 질의 언어와 무관하게 영어 지침을 내부 추론용으로 사용하고, 사용자 응답은 필요 시 다국어로 생성한다.
+
+---
+
+## Main Tool Files
+
+- MCP tool definitions: `src/mcp_airflow_api/airflow_api.py`
+- Utility functions: `src/mcp_airflow_api/functions.py`
+
+---
+
+## How To Use
+
+1. In your MCP Tools environment, configure `mcp-config.json` as follows:
+
+```json
+{
+	"mcpServers": {
+		"airflow-api": {
+			"command": "uvx",
+			"args": ["--python", "3.11", "mcp-airflow-api"],
+			"env": {
+				"AIRFLOW_API_URL": "http://localhost:38080/api/v1",
+				"AIRFLOW_API_USERNAME": "airflow",
+				"AIRFLOW_API_PASSWORD": "airflow",
+				"AIRFLOW_LOG_LEVEL": "INFO"
+			}
+		}
+	}
+}
+```
+
+2. Register the MCP server in MCP Inspector, OpenWebUI, Smithery, etc. and use the tools.
+
+---
+
+## QuickStart (Demo): Running MCP-Airflow-API with Docker
+
+1. Prepare an Airflow cluster  
+	 - See [Official Airflow Docker Install Guide](https://airflow.apache.org/docs/apache-airflow/stable/start/docker.html)
+
+2. Prepare MCP Tools environment  
+	 - Install Docker and Docker Compose
+	 - Clone this project and run `docker-compose up -d` in the root directory
+
+3. Register the MCP server in MCP Inspector/Smithery  
+	 - Example address: `http://localhost:8000/airflow-api`
+
+---
+
+## Logging & Observability
+
+- Structured logs for all tool invocations and HTTP requests
+- Control log level via environment variable (`AIRFLOW_LOG_LEVEL`) or CLI flag (`--log-level`)
+- Supported levels: DEBUG, INFO, WARNING, ERROR, CRITICAL
+
+---
+
+## License
+
+This project is licensed under the MIT License.
+
+---
+
+## Roadmap
+
+This project starts with a minimal set of essential Airflow management tools. Many more useful features and tools for Airflow cluster operations will be added soon, including advanced monitoring, DAG/task analytics, scheduling controls, and more. Contributions and suggestions are welcome!
+
+---
+
+## Additional Links
+
+- [Code](https://github.com/call518/MCP-Airflow-API)
+- [Issues](https://github.com/call518/MCP-Airflow-API/issues)
+- [Smithery Deployment](https://smithery.ai/server/@call518/mcp-airflow-api)
+

mcp_airflow_api-0.0.0/README.md

@@ -0,0 +1,156 @@
+Model Context Protocol (MCP) server for Apache Airflow API integration.  
+This project provides natural language MCP tools for essential Airflow cluster operations.
+
+[![Deploy to PyPI with tag](https://github.com/call518/MCP-Airflow-API/actions/workflows/pypi-publish.yml/badge.svg)](https://github.com/call518/MCP-Airflow-API/actions/workflows/pypi-publish.yml)
+
+[![smithery badge](https://smithery.ai/badge/@call518/mcp-airflow-api)](https://smithery.ai/server/@call518/mcp-airflow-api)
+
+---
+
+
+# MCP-Airflow-API
+
+**Tested and supported Airflow version: 2.10.2 (API Version: v1)**
+
+## Features
+
+- List all DAGs in the Airflow cluster
+- Monitor running/failed DAG runs
+- Trigger DAG runs on demand
+- Minimal, LLM-friendly output for all tools
+- Easy integration with MCP Inspector, OpenWebUI, Smithery, etc.
+
+---
+
+## Available MCP Tools
+
+### DAG Management
+
+- `list_dags`  
+	Returns all DAGs registered in the Airflow cluster.  
+	Output: `dag_id`, `dag_display_name`, `is_active`, `is_paused`, `owners`, `tags`
+
+- `running_dags`  
+	Returns all currently running DAG runs.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `failed_dags`  
+	Returns all recently failed DAG runs.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `trigger_dag(dag_id)`  
+	Immediately triggers the specified DAG.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `pause_dag(dag_id)`  
+	Pauses the specified DAG (prevents scheduling new runs).  
+	Output: `dag_id`, `is_paused`
+
+- `unpause_dag(dag_id)`  
+	Unpauses the specified DAG (allows scheduling new runs).  
+	Output: `dag_id`, `is_paused`
+
+---
+
+## Prompt Template
+
+The package exposes a tool `get_prompt_template` that returns either the entire template, a specific section, or just the headings. Three MCP prompts (`prompt_template_full`, `prompt_template_headings`, `prompt_template_section`) are also registered for discovery.
+
+### MCP Prompts
+
+For easier discoverability in MCP clients (so `prompts/list` is not empty), the server now registers three prompts:
+
+• `prompt_template_full` – returns the full canonical template  
+• `prompt_template_headings` – returns only the section headings  
+• `prompt_template_section` – takes a `section` argument (number or keyword) and returns that section
+
+You can still use the `get_prompt_template` tool for programmatic access or when you prefer tool invocation over prompt retrieval.
+
+Single canonical English prompt template guides safe and efficient tool selection.
+
+Files:
+• Packaged: `src/mcp_airflow_api/prompt_template.md` (distributed with PyPI)  
+• (Optional workspace root copy `PROMPT_TEMPLATE.md` may exist for editing; packaged copy is the one loaded at runtime.)
+
+Retrieve dynamically via MCP tool:
+• `get_prompt_template()` – full template  
+• `get_prompt_template("tool map")` – only the tool mapping section  
+• `get_prompt_template("3")` – section 3 (tool map)  
+• `get_prompt_template(mode="headings")` – list all section headings
+
+Policy: Only English is stored; LLM는 사용자 질의 언어와 무관하게 영어 지침을 내부 추론용으로 사용하고, 사용자 응답은 필요 시 다국어로 생성한다.
+
+---
+
+## Main Tool Files
+
+- MCP tool definitions: `src/mcp_airflow_api/airflow_api.py`
+- Utility functions: `src/mcp_airflow_api/functions.py`
+
+---
+
+## How To Use
+
+1. In your MCP Tools environment, configure `mcp-config.json` as follows:
+
+```json
+{
+	"mcpServers": {
+		"airflow-api": {
+			"command": "uvx",
+			"args": ["--python", "3.11", "mcp-airflow-api"],
+			"env": {
+				"AIRFLOW_API_URL": "http://localhost:38080/api/v1",
+				"AIRFLOW_API_USERNAME": "airflow",
+				"AIRFLOW_API_PASSWORD": "airflow",
+				"AIRFLOW_LOG_LEVEL": "INFO"
+			}
+		}
+	}
+}
+```
+
+2. Register the MCP server in MCP Inspector, OpenWebUI, Smithery, etc. and use the tools.
+
+---
+
+## QuickStart (Demo): Running MCP-Airflow-API with Docker
+
+1. Prepare an Airflow cluster  
+	 - See [Official Airflow Docker Install Guide](https://airflow.apache.org/docs/apache-airflow/stable/start/docker.html)
+
+2. Prepare MCP Tools environment  
+	 - Install Docker and Docker Compose
+	 - Clone this project and run `docker-compose up -d` in the root directory
+
+3. Register the MCP server in MCP Inspector/Smithery  
+	 - Example address: `http://localhost:8000/airflow-api`
+
+---
+
+## Logging & Observability
+
+- Structured logs for all tool invocations and HTTP requests
+- Control log level via environment variable (`AIRFLOW_LOG_LEVEL`) or CLI flag (`--log-level`)
+- Supported levels: DEBUG, INFO, WARNING, ERROR, CRITICAL
+
+---
+
+## License
+
+This project is licensed under the MIT License.
+
+---
+
+## Roadmap
+
+This project starts with a minimal set of essential Airflow management tools. Many more useful features and tools for Airflow cluster operations will be added soon, including advanced monitoring, DAG/task analytics, scheduling controls, and more. Contributions and suggestions are welcome!
+
+---
+
+## Additional Links
+
+- [Code](https://github.com/call518/MCP-Airflow-API)
+- [Issues](https://github.com/call518/MCP-Airflow-API/issues)
+- [Smithery Deployment](https://smithery.ai/server/@call518/mcp-airflow-api)
+

mcp_airflow_api-0.0.0/pyproject.toml

@@ -0,0 +1,25 @@
+[project]
+name = "mcp-airflow-api"
+version = "0.0.0"
+description = "Model Context Protocol (MCP) server for Apache Airflow API integration. Provides comprehensive tools for managing Airflow clusters including service operations, configuration management, status monitoring, and request tracking."
+readme = "README.md"
+requires-python = ">=3.11"
+dependencies = [
+    "fastapi>=0.116.1",
+    "requests>=2.32.4",
+    "uvicorn>=0.35.0",
+    "mcp>=1.12.3",
+]
+
+[project.scripts]
+mcp-airflow-api = "mcp_airflow_api.airflow_api:main"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+include = ["mcp_airflow_api*"]
+
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.package-data]
+mcp_airflow_api = ["prompt_template.md"]

mcp_airflow_api-0.0.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

mcp_airflow_api-0.0.0/src/mcp_airflow_api/__init__.py

@@ -0,0 +1 @@
+"""MCP Airflow API package."""

mcp_airflow_api-0.0.0/src/mcp_airflow_api/airflow_api.py

@@ -0,0 +1,270 @@
+"""
+MCP tool definitions for Airflow REST API operations.
+"""
+import argparse
+import logging
+from typing import Any, Dict, List, Optional
+import mcp
+from mcp.server.fastmcp import FastMCP
+import os
+from .functions import airflow_request, read_prompt_template, parse_prompt_sections
+
+# Setup logging
+logger = logging.getLogger(__name__)
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+)
+
+
+# MCP server instance for registering tools
+mcp = FastMCP("mcp-airflow-api")
+
+PROMPT_TEMPLATE_PATH = os.path.join(os.path.dirname(__file__), "prompt_template.md")
+
+
+
+@mcp.tool()
+def get_prompt_template(section: Optional[str] = None, mode: Optional[str] = None) -> str:
+    """
+    Returns the MCP prompt template (full, headings, or specific section).
+    Args:
+        section: Section number or keyword (optional)
+        mode: 'full', 'headings', or None (optional)
+    """
+    template = read_prompt_template(PROMPT_TEMPLATE_PATH)
+    
+    if mode == "headings":
+        headings, _ = parse_prompt_sections(template)
+        lines = ["Section Headings:"]
+        for idx, title in enumerate(headings, 1):
+            lines.append(f"{idx}. {title}")
+        return "\n".join(lines)
+    
+    if section:
+        headings, sections = parse_prompt_sections(template)
+        # Try by number
+        try:
+            idx = int(section) - 1
+            if 0 <= idx < len(sections):
+                return sections[idx]
+        except Exception:
+            pass
+        # Try by keyword
+        section_lower = section.strip().lower()
+        for i, heading in enumerate(headings):
+            if section_lower in heading.lower():
+                return sections[i]
+        return f"Section '{section}' not found."
+    
+    return template
+
+@mcp.tool()
+def list_dags() -> Dict[str, Any]:
+    """
+    [Tool Role]: Lists all DAGs registered in the Airflow cluster.
+
+    Returns:
+        List of DAGs with minimal info: dag_id, dag_display_name, is_active, is_paused, owners, tags
+    """
+    resp = airflow_request("GET", "/dags")
+    resp.raise_for_status()
+    dags = resp.json().get("dags", [])
+    minimal = []
+    for dag in dags:
+        minimal.append({
+            "dag_id": dag.get("dag_id"),
+            "dag_display_name": dag.get("dag_display_name"),
+            "is_active": dag.get("is_active"),
+            "is_paused": dag.get("is_paused"),
+            "owners": dag.get("owners"),
+            "tags": [t.get("name") for t in dag.get("tags", [])]
+        })
+    return {"dags": minimal}
+
+@mcp.tool()
+def running_dags() -> Dict[str, Any]:
+    """
+    [Tool Role]: Lists all currently running DAG runs in the Airflow cluster.
+
+    Returns:
+        List of running DAG runs with minimal info: dag_id, run_id, state, execution_date, start_date, end_date
+    """
+    dags_resp = airflow_request("GET", "/dags")
+    dags_resp.raise_for_status()
+    dags = dags_resp.json().get("dags", [])
+    running = []
+    for dag in dags:
+        dag_id = dag.get("dag_id")
+        if not dag_id:
+            continue
+        runs_resp = airflow_request("GET", f"/dags/{dag_id}/dagRuns")
+        runs_resp.raise_for_status()
+        runs = runs_resp.json().get("dag_runs", [])
+        for run in runs:
+            if run.get("state") == "running":
+                running.append({
+                    "dag_id": dag_id,
+                    "run_id": run.get("run_id"),
+                    "state": run.get("state"),
+                    "execution_date": run.get("execution_date"),
+                    "start_date": run.get("start_date"),
+                    "end_date": run.get("end_date")
+                })
+    return {"dag_runs": running}
+
+@mcp.tool()
+def failed_dags() -> Dict[str, Any]:
+    """
+    [Tool Role]: Lists all recently failed DAG runs in the Airflow cluster.
+
+    Returns:
+        List of failed DAG runs with minimal info: dag_id, run_id, state, execution_date, start_date, end_date
+    """
+    dags_resp = airflow_request("GET", "/dags")
+    dags_resp.raise_for_status()
+    dags = dags_resp.json().get("dags", [])
+    failed = []
+    for dag in dags:
+        dag_id = dag.get("dag_id")
+        if not dag_id:
+            continue
+        runs_resp = airflow_request("GET", f"/dags/{dag_id}/dagRuns")
+        runs_resp.raise_for_status()
+        runs = runs_resp.json().get("dag_runs", [])
+        for run in runs:
+            if run.get("state") == "failed":
+                failed.append({
+                    "dag_id": dag_id,
+                    "run_id": run.get("run_id"),
+                    "state": run.get("state"),
+                    "execution_date": run.get("execution_date"),
+                    "start_date": run.get("start_date"),
+                    "end_date": run.get("end_date")
+                })
+    return {"dag_runs": failed}
+
+@mcp.tool()
+def trigger_dag(dag_id: str) -> Dict[str, Any]:
+    """
+    [Tool Role]: Triggers a new DAG run for a specified Airflow DAG.
+
+    Args:
+        dag_id: The DAG ID to trigger
+
+    Returns:
+        Minimal info about triggered DAG run: dag_id, run_id, state, execution_date, start_date, end_date
+    """
+    if not dag_id:
+        raise ValueError("dag_id must not be empty")
+    resp = airflow_request("POST", f"/dags/{dag_id}/dagRuns", json={"conf": {}})
+    resp.raise_for_status()
+    run = resp.json()
+    return {
+        "dag_id": dag_id,
+        "run_id": run.get("run_id"),
+        "state": run.get("state"),
+        "execution_date": run.get("execution_date"),
+        "start_date": run.get("start_date"),
+        "end_date": run.get("end_date")
+    }
+
+@mcp.tool()
+def pause_dag(dag_id: str) -> Dict[str, Any]:
+    """
+    [Tool Role]: Pauses the specified Airflow DAG (prevents scheduling new runs).
+
+    Args:
+        dag_id: The DAG ID to pause
+
+    Returns:
+        Minimal info about the paused DAG: dag_id, is_paused
+    """
+    if not dag_id:
+        raise ValueError("dag_id must not be empty")
+    resp = airflow_request("PATCH", f"/dags/{dag_id}", json={"is_paused": True})
+    resp.raise_for_status()
+    dag = resp.json()
+    return {"dag_id": dag.get("dag_id", dag_id), "is_paused": dag.get("is_paused", True)}
+
+@mcp.tool()
+def unpause_dag(dag_id: str) -> Dict[str, Any]:
+    """
+    [Tool Role]: Unpauses the specified Airflow DAG (allows scheduling new runs).
+
+    Args:
+        dag_id: The DAG ID to unpause
+
+    Returns:
+        Minimal info about the unpaused DAG: dag_id, is_paused
+    """
+    if not dag_id:
+        raise ValueError("dag_id must not be empty")
+    resp = airflow_request("PATCH", f"/dags/{dag_id}", json={"is_paused": False})
+    resp.raise_for_status()
+    dag = resp.json()
+    return {"dag_id": dag.get("dag_id", dag_id), "is_paused": dag.get("is_paused", False)}
+
+#========================================================================================
+# MCP Prompts (for prompts/list exposure)
+#========================================================================================
+
+@mcp.prompt("prompt_template_full")
+def prompt_template_full_prompt() -> str:
+    """Return the full canonical prompt template."""
+    return read_prompt_template(PROMPT_TEMPLATE_PATH)
+
+@mcp.prompt("prompt_template_headings")
+def prompt_template_headings_prompt() -> str:
+    """Return compact list of section headings."""
+    template = read_prompt_template(PROMPT_TEMPLATE_PATH)
+    headings, _ = parse_prompt_sections(template)
+    lines = ["Section Headings:"]
+    for idx, title in enumerate(headings, 1):
+        lines.append(f"{idx}. {title}")
+    return "\n".join(lines)
+
+@mcp.prompt("prompt_template_section")
+def prompt_template_section_prompt(section: Optional[str] = None) -> str:
+    """Return a specific prompt template section by number or keyword."""
+    if not section:
+        headings_result = prompt_template_headings_prompt()
+        return "\n".join([
+            "[HELP] Missing 'section' argument.",
+            "Specify a section number or keyword.",
+            "Examples: 1 | overview | tool map | usage",
+            headings_result.strip()
+        ])
+    return get_prompt_template(section=section)
+
+#========================================================================================
+
+def main(argv: Optional[List[str]] = None):
+    """Entrypoint for MCP Airflow API server.
+
+    Supports optional CLI arguments (e.g. --log-level DEBUG) while remaining
+    backward-compatible with stdio launcher expectations.
+    """
+    parser = argparse.ArgumentParser(prog="mcp-airflow-api", description="MCP Airflow API Server")
+    parser.add_argument(
+        "--log-level", "-l",
+        dest="log_level",
+        help="Logging level override (DEBUG, INFO, WARNING, ERROR, CRITICAL). Overrides AIRFLOW_LOG_LEVEL env if provided.",
+        choices=["DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL"],
+    )
+    # Allow future extension without breaking unknown args usage
+    args = parser.parse_args(argv)
+
+    if args.log_level:
+        # Override root + specific logger level
+        logging.getLogger().setLevel(args.log_level)
+        logger.setLevel(args.log_level)
+        logging.getLogger("requests.packages.urllib3").setLevel("WARNING")  # reduce noise at DEBUG
+        logger.info("Log level set via CLI to %s", args.log_level)
+    else:
+        logger.debug("Log level from environment: %s", logging.getLogger().level)
+
+    mcp.run(transport='stdio')
+
+if __name__ == "__main__":
+    main()

mcp_airflow_api-0.0.0/src/mcp_airflow_api/functions.py

@@ -0,0 +1,63 @@
+"""
+Utility functions for Airflow MCP
+"""
+import os
+import requests
+from typing import Any, Dict, Optional
+
+def airflow_request(method: str, path: str, **kwargs) -> requests.Response:
+    """
+    Make a Basic Auth request to Airflow REST API.
+    'path' should be relative to AIRFLOW_API_URL (e.g., '/dags', '/pools').
+    """
+    base_url = os.getenv("AIRFLOW_API_URL", "").rstrip("/")
+    if not base_url:
+        raise RuntimeError("AIRFLOW_API_URL environment variable is not set")
+    
+    # Ensure path starts with /
+    if not path.startswith("/"):
+        path = "/" + path
+    
+    # Construct full URL
+    full_url = base_url + path
+    
+    # Get authentication
+    username = os.getenv("AIRFLOW_API_USERNAME")
+    password = os.getenv("AIRFLOW_API_PASSWORD")
+    if not username or not password:
+        raise RuntimeError("AIRFLOW_API_USERNAME or AIRFLOW_API_PASSWORD environment variable is not set")
+    
+    auth = (username, password)
+    headers = kwargs.pop("headers", {})
+    
+    return requests.request(method, full_url, headers=headers, auth=auth, **kwargs)
+
+def read_prompt_template(path: str) -> str:
+    """
+    Reads the MCP prompt template file and returns its content as a string.
+    """
+    with open(path, "r", encoding="utf-8") as f:
+        return f.read()
+
+
+def parse_prompt_sections(template: str):
+    """
+    Parses the prompt template into section headings and sections.
+    Returns (headings, sections).
+    """
+    lines = template.splitlines()
+    sections = []
+    current = []
+    headings = []
+    for line in lines:
+        if line.startswith("## "):
+            if current:
+                sections.append("\n".join(current))
+                current = []
+            headings.append(line[3:].strip())
+            current.append(line)
+        else:
+            current.append(line)
+    if current:
+        sections.append("\n".join(current))
+    return headings, sections

mcp_airflow_api-0.0.0/src/mcp_airflow_api/prompt_template.md

@@ -0,0 +1,56 @@
+# MCP Airflow API Prompt Template
+
+## 1. Overview
+
+This MCP server provides natural language tools for managing Apache Airflow clusters via REST API. All prompts and tool outputs are designed for minimal, LLM-friendly English responses.
+
+## 2. Available MCP Tools
+
+- `list_dags`: List all DAGs in the Airflow cluster.
+- `running_dags`: List all currently running DAG runs.
+- `failed_dags`: List all recently failed DAG runs.
+- `trigger_dag(dag_id)`: Trigger a DAG run by ID.
+- `pause_dag(dag_id)`: Pause a DAG (prevent scheduling).
+- `unpause_dag(dag_id)`: Unpause a DAG (allow scheduling).
+
+## 3. Tool Map
+
+| Tool Name      | Role/Description                          | Input Args      | Output Fields                        |
+|----------------|-------------------------------------------|-----------------|--------------------------------------|
+| list_dags      | List all DAGs                             | None            | dag_id, dag_display_name, is_active, is_paused, owners, tags |
+| running_dags   | List running DAG runs                     | None            | dag_id, run_id, state, execution_date, start_date, end_date |
+| failed_dags    | List failed DAG runs                      | None            | dag_id, run_id, state, execution_date, start_date, end_date |
+| trigger_dag    | Trigger a DAG run                         | dag_id (str)    | dag_id, run_id, state, execution_date, start_date, end_date |
+| pause_dag      | Pause a DAG                               | dag_id (str)    | dag_id, is_paused                    |
+| unpause_dag    | Unpause a DAG                             | dag_id (str)    | dag_id, is_paused                    |
+
+## 4. Usage Guidelines
+
+- Always use minimal, structured output.
+- All tool invocations must use English for internal reasoning.
+- For user-facing responses, translate to the user's language if needed.
+
+## 5. Example Queries
+
+- "List all DAGs."
+- "Show running DAGs."
+- "Trigger DAG 'example_dag'."
+- "Pause DAG 'etl_job'."
+- "Unpause DAG 'etl_job'."
+
+## 6. Formatting Rules
+
+- Output only the requested fields.
+- No extra explanation unless explicitly requested.
+- Use JSON objects for tool outputs.
+
+## 7. Logging & Environment
+
+- Control log level via AIRFLOW_LOG_LEVEL env or --log-level CLI flag.
+- Supported levels: DEBUG, INFO, WARNING, ERROR, CRITICAL.
+
+## 8. References
+
+- Main MCP tool file: `src/mcp_airflow_api/airflow_api.py`
+- Utility functions: `src/mcp_airflow_api/functions.py`
+- See README.md for full usage and configuration.

mcp_airflow_api-0.0.0/src/mcp_airflow_api.egg-info/PKG-INFO

@@ -0,0 +1,169 @@
+Metadata-Version: 2.4
+Name: mcp-airflow-api
+Version: 0.0.0
+Summary: Model Context Protocol (MCP) server for Apache Airflow API integration. Provides comprehensive tools for managing Airflow clusters including service operations, configuration management, status monitoring, and request tracking.
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: fastapi>=0.116.1
+Requires-Dist: requests>=2.32.4
+Requires-Dist: uvicorn>=0.35.0
+Requires-Dist: mcp>=1.12.3
+Dynamic: license-file
+
+Model Context Protocol (MCP) server for Apache Airflow API integration.  
+This project provides natural language MCP tools for essential Airflow cluster operations.
+
+[![Deploy to PyPI with tag](https://github.com/call518/MCP-Airflow-API/actions/workflows/pypi-publish.yml/badge.svg)](https://github.com/call518/MCP-Airflow-API/actions/workflows/pypi-publish.yml)
+
+[![smithery badge](https://smithery.ai/badge/@call518/mcp-airflow-api)](https://smithery.ai/server/@call518/mcp-airflow-api)
+
+---
+
+
+# MCP-Airflow-API
+
+**Tested and supported Airflow version: 2.10.2 (API Version: v1)**
+
+## Features
+
+- List all DAGs in the Airflow cluster
+- Monitor running/failed DAG runs
+- Trigger DAG runs on demand
+- Minimal, LLM-friendly output for all tools
+- Easy integration with MCP Inspector, OpenWebUI, Smithery, etc.
+
+---
+
+## Available MCP Tools
+
+### DAG Management
+
+- `list_dags`  
+	Returns all DAGs registered in the Airflow cluster.  
+	Output: `dag_id`, `dag_display_name`, `is_active`, `is_paused`, `owners`, `tags`
+
+- `running_dags`  
+	Returns all currently running DAG runs.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `failed_dags`  
+	Returns all recently failed DAG runs.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `trigger_dag(dag_id)`  
+	Immediately triggers the specified DAG.  
+	Output: `dag_id`, `run_id`, `state`, `execution_date`, `start_date`, `end_date`
+
+- `pause_dag(dag_id)`  
+	Pauses the specified DAG (prevents scheduling new runs).  
+	Output: `dag_id`, `is_paused`
+
+- `unpause_dag(dag_id)`  
+	Unpauses the specified DAG (allows scheduling new runs).  
+	Output: `dag_id`, `is_paused`
+
+---
+
+## Prompt Template
+
+The package exposes a tool `get_prompt_template` that returns either the entire template, a specific section, or just the headings. Three MCP prompts (`prompt_template_full`, `prompt_template_headings`, `prompt_template_section`) are also registered for discovery.
+
+### MCP Prompts
+
+For easier discoverability in MCP clients (so `prompts/list` is not empty), the server now registers three prompts:
+
+• `prompt_template_full` – returns the full canonical template  
+• `prompt_template_headings` – returns only the section headings  
+• `prompt_template_section` – takes a `section` argument (number or keyword) and returns that section
+
+You can still use the `get_prompt_template` tool for programmatic access or when you prefer tool invocation over prompt retrieval.
+
+Single canonical English prompt template guides safe and efficient tool selection.
+
+Files:
+• Packaged: `src/mcp_airflow_api/prompt_template.md` (distributed with PyPI)  
+• (Optional workspace root copy `PROMPT_TEMPLATE.md` may exist for editing; packaged copy is the one loaded at runtime.)
+
+Retrieve dynamically via MCP tool:
+• `get_prompt_template()` – full template  
+• `get_prompt_template("tool map")` – only the tool mapping section  
+• `get_prompt_template("3")` – section 3 (tool map)  
+• `get_prompt_template(mode="headings")` – list all section headings
+
+Policy: Only English is stored; LLM는 사용자 질의 언어와 무관하게 영어 지침을 내부 추론용으로 사용하고, 사용자 응답은 필요 시 다국어로 생성한다.
+
+---
+
+## Main Tool Files
+
+- MCP tool definitions: `src/mcp_airflow_api/airflow_api.py`
+- Utility functions: `src/mcp_airflow_api/functions.py`
+
+---
+
+## How To Use
+
+1. In your MCP Tools environment, configure `mcp-config.json` as follows:
+
+```json
+{
+	"mcpServers": {
+		"airflow-api": {
+			"command": "uvx",
+			"args": ["--python", "3.11", "mcp-airflow-api"],
+			"env": {
+				"AIRFLOW_API_URL": "http://localhost:38080/api/v1",
+				"AIRFLOW_API_USERNAME": "airflow",
+				"AIRFLOW_API_PASSWORD": "airflow",
+				"AIRFLOW_LOG_LEVEL": "INFO"
+			}
+		}
+	}
+}
+```
+
+2. Register the MCP server in MCP Inspector, OpenWebUI, Smithery, etc. and use the tools.
+
+---
+
+## QuickStart (Demo): Running MCP-Airflow-API with Docker
+
+1. Prepare an Airflow cluster  
+	 - See [Official Airflow Docker Install Guide](https://airflow.apache.org/docs/apache-airflow/stable/start/docker.html)
+
+2. Prepare MCP Tools environment  
+	 - Install Docker and Docker Compose
+	 - Clone this project and run `docker-compose up -d` in the root directory
+
+3. Register the MCP server in MCP Inspector/Smithery  
+	 - Example address: `http://localhost:8000/airflow-api`
+
+---
+
+## Logging & Observability
+
+- Structured logs for all tool invocations and HTTP requests
+- Control log level via environment variable (`AIRFLOW_LOG_LEVEL`) or CLI flag (`--log-level`)
+- Supported levels: DEBUG, INFO, WARNING, ERROR, CRITICAL
+
+---
+
+## License
+
+This project is licensed under the MIT License.
+
+---
+
+## Roadmap
+
+This project starts with a minimal set of essential Airflow management tools. Many more useful features and tools for Airflow cluster operations will be added soon, including advanced monitoring, DAG/task analytics, scheduling controls, and more. Contributions and suggestions are welcome!
+
+---
+
+## Additional Links
+
+- [Code](https://github.com/call518/MCP-Airflow-API)
+- [Issues](https://github.com/call518/MCP-Airflow-API/issues)
+- [Smithery Deployment](https://smithery.ai/server/@call518/mcp-airflow-api)
+

mcp_airflow_api-0.0.0/src/mcp_airflow_api.egg-info/SOURCES.txt

@@ -0,0 +1,14 @@
+LICENSE
+MANIFEST.in
+README.md
+pyproject.toml
+src/mcp_airflow_api/__init__.py
+src/mcp_airflow_api/airflow_api.py
+src/mcp_airflow_api/functions.py
+src/mcp_airflow_api/prompt_template.md
+src/mcp_airflow_api.egg-info/PKG-INFO
+src/mcp_airflow_api.egg-info/SOURCES.txt
+src/mcp_airflow_api.egg-info/dependency_links.txt
+src/mcp_airflow_api.egg-info/entry_points.txt
+src/mcp_airflow_api.egg-info/requires.txt
+src/mcp_airflow_api.egg-info/top_level.txt

mcp_airflow_api-0.0.0/src/mcp_airflow_api.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

mcp_airflow_api-0.0.0/src/mcp_airflow_api.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+mcp-airflow-api = mcp_airflow_api.airflow_api:main

mcp_airflow_api-0.0.0/src/mcp_airflow_api.egg-info/requires.txt

@@ -0,0 +1,4 @@
+fastapi>=0.116.1
+requests>=2.32.4
+uvicorn>=0.35.0
+mcp>=1.12.3

mcp_airflow_api-0.0.0/src/mcp_airflow_api.egg-info/top_level.txt

@@ -0,0 +1 @@
+mcp_airflow_api