Skip to main content
Glama
nikhil-ganage

MCP Server Airflow Token

set_task_instances_state

Change the state of Airflow task instances to manage workflow execution, control task behavior, and resolve issues in DAG runs.

Instructions

Set a state of task instances

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
dag_idYes
stateYes
task_idsNo
execution_dateNo
include_upstreamNo
include_downstreamNo
include_futureNo
include_pastNo
dry_runNo

Implementation Reference

  • The async handler function that executes the tool's logic by building a state request and calling the Airflow DAG API's post_set_task_instances_state method.
    async def set_task_instances_state(
        dag_id: str,
        state: str,
        task_ids: Optional[List[str]] = None,
        execution_date: Optional[str] = None,
        include_upstream: Optional[bool] = None,
        include_downstream: Optional[bool] = None,
        include_future: Optional[bool] = None,
        include_past: Optional[bool] = None,
        dry_run: Optional[bool] = None,
    ) -> List[Union[types.TextContent, types.ImageContent, types.EmbeddedResource]]:
        state_request = {"state": state}
        if task_ids is not None:
            state_request["task_ids"] = task_ids
        if execution_date is not None:
            state_request["execution_date"] = execution_date
        if include_upstream is not None:
            state_request["include_upstream"] = include_upstream
        if include_downstream is not None:
            state_request["include_downstream"] = include_downstream
        if include_future is not None:
            state_request["include_future"] = include_future
        if include_past is not None:
            state_request["include_past"] = include_past
        if dry_run is not None:
            state_request["dry_run"] = dry_run
    
        update_task_instances_state = UpdateTaskInstancesState(**state_request)
    
        response = dag_api.post_set_task_instances_state(
            dag_id=dag_id,
            update_task_instances_state=update_task_instances_state,
        )
        return [types.TextContent(type="text", text=str(response.to_dict()))]
  • The get_all_functions() defines the list of tools from the DAG module, including the registration tuple for set_task_instances_state.
    def get_all_functions() -> list[tuple[Callable, str, str, bool]]:
        """Return list of (function, name, description, is_read_only) tuples for registration."""
        return [
            (get_dags, "fetch_dags", "Fetch all DAGs", True),
            (get_dag, "get_dag", "Get a DAG by ID", True),
            (get_dag_details, "get_dag_details", "Get a simplified representation of DAG", True),
            (get_dag_source, "get_dag_source", "Get a source code", True),
            (pause_dag, "pause_dag", "Pause a DAG by ID", False),
            (unpause_dag, "unpause_dag", "Unpause a DAG by ID", False),
            (get_dag_tasks, "get_dag_tasks", "Get tasks for DAG", True),
            (get_task, "get_task", "Get a task by ID", True),
            (get_tasks, "get_tasks", "Get tasks for DAG", True),
            (patch_dag, "patch_dag", "Update a DAG", False),
            (patch_dags, "patch_dags", "Update multiple DAGs", False),
            (delete_dag, "delete_dag", "Delete a DAG", False),
            (clear_task_instances, "clear_task_instances", "Clear a set of task instances", False),
            (set_task_instances_state, "set_task_instances_state", "Set a state of task instances", False),
            (reparse_dag_file, "reparse_dag_file", "Request re-parsing of a DAG file", False),
        ]
  • src/main.py:78-98 (registration)
    The main.py script iterates over API modules, calls their get_all_functions(), and registers each tool with the MCP app using app.add_tool().
    for api in apis:
        logging.debug(f"Adding API: {api}")
        get_function = APITYPE_TO_FUNCTIONS[APIType(api)]
        try:
            functions = get_function()
        except NotImplementedError:
            continue
    
        # Filter functions for read-only mode if requested
        if read_only:
            functions = filter_functions_for_read_only(functions)
    
        for func, name, description, *_ in functions:
            app.add_tool(func, name=name, description=description)
    
    if transport == "sse":
        logging.debug("Starting MCP server for Apache Airflow with SSE transport")
        app.run(transport="sse")
    else:
        logging.debug("Starting MCP server for Apache Airflow with stdio transport")
        app.run(transport="stdio")
Behavior1/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure but provides almost none. 'Set a state' implies a mutation/write operation, but the description doesn't mention permissions required, whether this is destructive or reversible, rate limits, side effects, or what happens to downstream/upstream dependencies. For a tool with 9 parameters that can modify workflow states, this is critically inadequate.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness2/5

Is the description appropriately sized, front-loaded, and free of redundancy?

While technically concise with just 5 words, this is under-specification rather than effective conciseness. The single sentence doesn't front-load critical information and fails to provide the necessary context for a complex tool. Every word earns its place, but there are far too few words for what this tool does.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness1/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with 9 parameters, no annotations, no output schema, and 0% schema description coverage, the description is completely inadequate. It doesn't explain what the tool actually does beyond the name, provides no behavioral context, offers no parameter guidance, and gives no indication of when to use it versus sibling tools. This leaves an AI agent with insufficient information to use the tool correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The description mentions 'state' but provides no context about what states are valid or what setting a state means. With 0% schema description coverage and 9 parameters (7 optional), the description fails to explain any parameters beyond the minimal implication of 'state' in the tool name. It doesn't clarify what 'dag_id', 'task_ids', 'execution_date', or the various include/dry_run parameters mean or how they interact.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose2/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description 'Set a state of task instances' is a tautology that essentially restates the tool name 'set_task_instances_state' with minimal additional information. It doesn't specify what kind of state (e.g., success, failed, running) or provide any meaningful differentiation from sibling tools like 'clear_task_instances' or 'update_task_instance'.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines1/5

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. There are multiple sibling tools that manipulate task instances (clear_task_instances, update_task_instance) and DAG states (update_dag_run_state), but the description offers no context about appropriate use cases, prerequisites, or distinctions from these related tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/nikhil-ganage/mcp-server-airflow-token'

If you have feedback or need assistance with the MCP directory API, please join our Discord server