Skip to main content
Glama

add_task

Create new tasks with optional Smart Add syntax for due dates, priorities, tags, locations, time estimates, and repetition. Parse natural language to set task details.

Instructions

Add a new task.

Supports Smart Add syntax when parse=True: - ^date for due date (^tomorrow, ^next friday) - !priority (!1, !2, !3) - #tag for tags (#work, #urgent) - @location - =time estimate (=30min, =1h) - *repeat pattern (*daily, *weekly)

Args: name: Task name (with optional Smart Add syntax) list_name: List to add to (uses default list if not specified) parse: Parse Smart Add syntax (default: True)

Returns: Created task details with transaction ID for undo

Examples: - add_task("Buy groceries") - add_task("Call mom ^tomorrow !1 #family") - add_task("Weekly review *weekly ^monday", list_name="Work")

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
nameYes
list_nameNo
parseNo

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • The add_task tool handler function. Decorated with @mcp.tool() inside register_task_tools. Accepts name, optional list_name, and parse flag. Makes rtm.tasks.add API call with Smart Add syntax support and returns formatted task response with transaction ID for undo.
    @mcp.tool()
    async def add_task(
        ctx: Context,
        name: str,
        list_name: str | None = None,
        parse: bool = True,
    ) -> dict[str, Any]:
        """Add a new task.
    
        Supports Smart Add syntax when parse=True:
            - ^date for due date (^tomorrow, ^next friday)
            - !priority (!1, !2, !3)
            - #tag for tags (#work, #urgent)
            - @location
            - =time estimate (=30min, =1h)
            - *repeat pattern (*daily, *weekly)
    
        Args:
            name: Task name (with optional Smart Add syntax)
            list_name: List to add to (uses default list if not specified)
            parse: Parse Smart Add syntax (default: True)
    
        Returns:
            Created task details with transaction ID for undo
    
        Examples:
            - add_task("Buy groceries")
            - add_task("Call mom ^tomorrow !1 #family")
            - add_task("Weekly review *weekly ^monday", list_name="Work")
        """
        client: RTMClient = await get_client()
    
        params: dict[str, Any] = {
            "name": name,
            "parse": "1" if parse else "0",
        }
    
        if list_name:
            lists_result = await client.call("rtm.lists.getList")
            from ..response_builder import parse_lists_response
    
            lists = parse_lists_response(lists_result)
            for lst in lists:
                if lst["name"].lower() == list_name.lower():
                    params["list_id"] = lst["id"]
                    break
    
        result = await client.call("rtm.tasks.add", require_timeline=True, **params)
    
        # Parse the created task
        tasks = parse_tasks_response(result)
        task = tasks[0] if tasks else {}
        transaction_id = get_transaction_id(result)
        timezone = await _get_user_timezone(client)
    
        return build_response(
            data={
                "task": format_task(task, timezone=timezone),
                "message": f"Created task: {task.get('name', name)}",
            },
            transaction_id=transaction_id,
        )
  • register_task_tools function that registers all task tools including add_task via @mcp.tool() decorator.
    def register_task_tools(mcp: Any, get_client: Any) -> None:
  • Server entry point calling register_task_tools(mcp, get_client) to register all task tools including add_task.
    # Register all tools
    register_task_tools(mcp, get_client)
    register_list_tools(mcp, get_client)
    register_note_tools(mcp, get_client)
    register_utility_tools(mcp, get_client)
  • Helper functions used by add_task: build_response for consistent response format, format_task for task display, parse_tasks_response for parsing RTM API response, get_transaction_id for undo support.
    def build_response(
        data: dict[str, Any] | list[Any],
        analysis: dict[str, Any] | None = None,
        transaction_id: str | None = None,
    ) -> dict[str, Any]:
        """Build a consistent response structure.
    
        Args:
            data: The main response data
            analysis: Optional analysis/insights
            transaction_id: Optional transaction ID for undo support
    
        Returns:
            Structured response dict
        """
        response = {
            "data": data,
            "metadata": {
                "fetched_at": datetime.now().isoformat(),
            },
        }
    
        if analysis:
            response["analysis"] = analysis
    
        if transaction_id:
            response["metadata"]["transaction_id"] = transaction_id
    
        return response
    
    
    def _convert_rtm_date(due: str, timezone: str | None) -> str:
        """Convert RTM date (UTC) to user's timezone.
    
        Args:
            due: Date string from RTM (ISO 8601 format, typically with Z suffix)
            timezone: User's IANA timezone (e.g., 'Europe/Warsaw')
    
        Returns:
            ISO 8601 date string in user's timezone, or original if conversion fails
        """
        if not timezone:
            return due
    
        try:
            from zoneinfo import ZoneInfo
    
            # Parse the UTC date from RTM
            due_dt = datetime.fromisoformat(due.replace("Z", "+00:00"))
    
            # Convert to user's timezone
            user_tz = ZoneInfo(timezone)
            due_local = due_dt.astimezone(user_tz)
    
            # Return ISO format in user's timezone
            return due_local.isoformat()
        except Exception:
            # If conversion fails, return original
            return due
    
    
    def format_task(
        task: dict[str, Any], include_ids: bool = True, timezone: str | None = None
    ) -> dict[str, Any]:
        """Format a task for response.
    
        Args:
            task: Raw task data from RTM
            include_ids: Whether to include task IDs
            timezone: User's IANA timezone for date conversion (e.g., 'Europe/Warsaw')
    
        Returns:
            Formatted task dict
        """
        # Convert dates to user's timezone
        due_display = None
        due_raw = task.get("due")
        if due_raw:
            due_display = _convert_rtm_date(due_raw, timezone)
    
        start_display = None
        start_raw = task.get("start")
        if start_raw:
            start_display = _convert_rtm_date(start_raw, timezone)
    
        formatted = {
            "name": task.get("name", ""),
            "priority": _priority_label(task.get("priority", "N")),
            "due": due_display,
            "start": start_display,
            "completed": task.get("completed") or None,
            "tags": task.get("tags", []),
            "url": task.get("url") or None,
            "notes_count": len(task.get("notes", [])),
            "estimate": task.get("estimate") or None,
            "modified": task.get("modified") or None,
        }
    
        if include_ids:
            formatted["id"] = task.get("id")
            formatted["taskseries_id"] = task.get("taskseries_id")
            formatted["list_id"] = task.get("list_id")
    
        return formatted
    
    
    def format_list(lst: dict[str, Any]) -> dict[str, Any]:
        """Format a list for response."""
        return {
            "id": lst.get("id"),
            "name": lst.get("name"),
            "smart": lst.get("smart") == "1",
            "archived": lst.get("archived") == "1",
            "locked": lst.get("locked") == "1",
        }
    
    
    def _priority_label(priority: str) -> str:
        """Convert priority code to label."""
        labels = {
            "1": "high",
            "2": "medium",
            "3": "low",
            "N": "none",
        }
        return labels.get(priority, "none")
    
    
    def priority_to_code(priority: str | int | None) -> str:
        """Convert priority label/number to RTM code."""
        if priority is None:
            return "N"
    
        priority_str = str(priority).lower()
    
        mapping = {
            "high": "1",
            "1": "1",
            "medium": "2",
            "2": "2",
            "low": "3",
            "3": "3",
            "none": "N",
            "0": "N",
            "n": "N",
        }
    
        return mapping.get(priority_str, "N")
    
    
    def parse_tasks_response(result: dict[str, Any]) -> list[dict[str, Any]]:
        """Parse RTM tasks response into flat task list.
    
        RTM returns nested structure:
        tasks.list[].taskseries[].task[]
    
        We flatten this to a simple list with all IDs attached.
        """
        tasks = []
        task_lists = result.get("tasks", {}).get("list", [])
    
        if isinstance(task_lists, dict):
            task_lists = [task_lists]
    
        for tl in task_lists:
            list_id = tl.get("id")
            taskseries_list = tl.get("taskseries", [])
    
            if isinstance(taskseries_list, dict):
                taskseries_list = [taskseries_list]
    
            for ts in taskseries_list:
                task_data = ts.get("task", [])
                if isinstance(task_data, dict):
                    task_data = [task_data]
    
                # Parse tags
                tags_data = ts.get("tags", [])
                if isinstance(tags_data, dict):
                    tags = tags_data.get("tag", [])
                    if isinstance(tags, str):
                        tags = [tags]
                else:
                    tags = []
    
                # Parse notes
                notes_data = ts.get("notes", [])
                notes = []
                if isinstance(notes_data, dict):
                    notes = notes_data.get("note", [])
                    if isinstance(notes, dict):
                        notes = [notes]
    
                for t in task_data:
                    tasks.append(
                        {
                            "id": t.get("id"),
                            "taskseries_id": ts.get("id"),
                            "list_id": list_id,
                            "name": ts.get("name"),
                            "due": t.get("due") or None,
                            "has_due_time": t.get("has_due_time") == "1",
                            "start": t.get("start") or None,
                            "has_start_time": t.get("has_start_time") == "1",
                            "completed": t.get("completed") or None,
                            "deleted": t.get("deleted") or None,
                            "priority": t.get("priority", "N"),
                            "postponed": int(t.get("postponed", 0)),
                            "estimate": t.get("estimate") or None,
                            "tags": tags if tags else [],
                            "notes": notes,
                            "url": ts.get("url") or None,
                            "location_id": ts.get("location_id") or None,
                            "created": ts.get("created") or None,
                            "modified": ts.get("modified") or None,
                        }
                    )
    
        return tasks
    
    
    def parse_lists_response(result: dict[str, Any]) -> list[dict[str, Any]]:
        """Parse RTM lists response."""
        lists = result.get("lists", {}).get("list", [])
        if isinstance(lists, dict):
            lists = [lists]
    
        return [
            {
                "id": lst.get("id"),
                "name": lst.get("name"),
                "deleted": lst.get("deleted") == "1",
                "locked": lst.get("locked") == "1",
                "archived": lst.get("archived") == "1",
                "position": int(lst.get("position", -1)),
                "smart": lst.get("smart") == "1",
                "filter": lst.get("filter"),
                "sort_order": lst.get("sort_order"),
            }
            for lst in lists
        ]
    
    
    def get_transaction_id(result: dict[str, Any]) -> str | None:
        """Extract transaction ID from response for undo support."""
        transaction = result.get("transaction", {})
        return transaction.get("id")
  • Input schema for add_task: name (str, required), list_name (str | None, optional), parse (bool, default True). Returns dict with task details and transaction_id.
    async def add_task(
        ctx: Context,
        name: str,
        list_name: str | None = None,
        parse: bool = True,
    ) -> dict[str, Any]:
        """Add a new task.
    
        Supports Smart Add syntax when parse=True:
            - ^date for due date (^tomorrow, ^next friday)
            - !priority (!1, !2, !3)
            - #tag for tags (#work, #urgent)
            - @location
            - =time estimate (=30min, =1h)
            - *repeat pattern (*daily, *weekly)
    
        Args:
            name: Task name (with optional Smart Add syntax)
            list_name: List to add to (uses default list if not specified)
            parse: Parse Smart Add syntax (default: True)
    
        Returns:
            Created task details with transaction ID for undo
    
        Examples:
            - add_task("Buy groceries")
            - add_task("Call mom ^tomorrow !1 #family")
            - add_task("Weekly review *weekly ^monday", list_name="Work")
        """
Behavior4/5

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

With no annotations, the description carries the full burden. It discloses creation behavior, Smart Add parsing, and return of transaction ID for undo. However, it omits potential side effects like non-idempotency or auth requirements.

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

Conciseness5/5

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

The description is well-structured with a summary, syntax table, args, returns, and examples. Every sentence serves a purpose without redundancy.

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

Completeness4/5

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

Given complexity (Smart Add, 3 params, output schema), the description covers essentials but misses prerequisites (e.g., authentication) and error scenarios. Output schema exists, so return details are not required here.

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

Parameters5/5

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

Schema coverage is 0%, but the description compensates fully by explaining each parameter (name, list_name, parse) with syntax details and examples, adding significant meaning beyond the bare schema.

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

Purpose5/5

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

The description clearly states 'Add a new task' and details Smart Add syntax, distinguishing it from sibling tools like add_note, add_list, etc. The examples further reinforce the purpose.

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

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies use for adding tasks, but lacks explicit guidance on when to use this tool versus alternatives like add_task_tags or set_task_due_date. No 'when not to use' or conditional context provided.

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/ljadach/rtm-mcp'

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