Skip to main content
Glama
partymola

ticktick-mcp

by partymola
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

ticktick_filter_tasks

Filter tasks in TickTick by status, project, priority, tag, and date ranges to retrieve matching task lists.

Instructions

Return the tasks matching every supplied filter criterion.

Supports any combination of project, priority, tag, status, and a date window applied to either the due date (open tasks) or the completion timestamp (completed tasks).

Args: detail (str, optional): "compact" (default) or "full". Compact drops the heavy content/desc/checklist items blobs and bulky sync metadata, keeping id, projectId, title, dueDate, startDate, priority, status, isAllDay, timeZone, tags plus a contentPreview (first ~200 chars of content) so keyword search still works. Full returns the raw task objects unchanged. To EDIT a task, fetch the full object with ticktick_get_by_id first, then send every field back via ticktick_update_task -- compact output must never feed an update. filter_criteria (dict | str): A criteria object, or a JSON string that decodes to one. No value below reaches the query unchecked, and any key not on this list is an error rather than something ignored, so a mistyped key fails loudly. Recognised keys:

    * ``status``: ``"uncompleted"`` (default) or ``"completed"``.
      When ``"completed"`` you should supply
      ``completion_start_date`` and/or ``completion_end_date``;
      without dates the result is an empty list.
    * ``project_id`` (str): Limit to tasks in this project.
      Accepts the project's name as well as its ID
      (case-insensitive, trimmed). Two projects sharing a
      name is an error, not a guess, and so is a project the
      account does not have.
    * ``priority`` (int): 0=None, 1=Low, 3=Medium, 5=High. Must be
      a JSON integer; the string ``"3"`` and the float ``3.0`` are
      errors, as is any other value.
    * ``tag_label`` (str): Tag name (case-sensitive), non-empty.
      Surrounding whitespace is trimmed.
    * ``due_start_date`` / ``due_end_date`` (str): ISO date or
      datetime strings; only used when ``status='uncompleted'``.
      A value that cannot be read as a date is an error, so a
      window is never silently dropped, and a window that ends
      before it begins is an error too. Bounds apply at day
      granularity, so any time part is ignored rather than
      validated.
    * ``completion_start_date`` / ``completion_end_date`` (str):
      ISO date or datetime strings; only used when
      ``status='completed'``. Same date rules as above.
    * ``tz`` (str): Default IANA timezone applied to date filters,
      e.g. ``"Europe/London"``. A name that does not resolve is an
      error rather than being ignored, so results are never
      labelled with a zone that was not applied.
    * ``sort_by_priority`` (bool): Sort by descending priority.
      Must be a JSON boolean; the strings ``"true"``/``"false"``
      and ``1``/``0`` are errors.

Returns: JSON list of matching task objects (compact by default; see detail). Empty list if nothing matches. If a compact result would still exceed the size budget, the soonest-due matches are returned and a final _truncation_note element reports how many were omitted -- nothing is dropped silently. On invalid input or backend failure: {"error": "...", "status": "error"}. One error carries an extra key: outcome: "project_list_unverifiable" means the project list could not be refreshed to confirm project_id, so retry rather than concluding the project is gone.

Freshness: Uncompleted queries read local state, synced from the server at most once per throttle window (default 15s, TICKTICK_MCP_SYNC_TTL_SECONDS); a change made elsewhere within that window may not be visible yet -- call ticktick_sync to force a refresh. Completed queries are always fetched live.

Limitations: - TickTick caps get_completed at 100 results; very wide completion windows are truncated server-side. - Filtering happens client-side after the fetch, so additional criteria do not reduce the number of network requests. - Compact output is for browsing only; full content for one task is available via ticktick_get_by_id or detail="full".

Agent Usage Guide: - List open tasks in a project: {"status": "uncompleted", "project_id": "<id>"} - List completed tasks in the last 7 days: { "status": "completed", "project_id": "<id>", "completion_start_date": "2026-05-21", "completion_end_date": "2026-05-28" } - Find high-priority open tasks due this month, sorted: { "priority": 5, "due_start_date": "2026-05-01", "due_end_date": "2026-05-31", "sort_by_priority": true }

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
detailNocompact
filter_criteriaYes

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.9/5.0
Behavior5/5

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

With no annotations provided, the description carries the full burden and meets it: it discloses local-vs-live data sources, throttle-window freshness, strict validation failures ("any key not on this list is an error"), truncation with a _truncation_note, the error envelope format, and the project_list_unverifiable retry case. Nothing is dropped silently and side effects are clearly absent.

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

Conciseness4/5

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

The description is long, but the length is justified by the bare schema and the tool's genuine complexity, and it is well organized with headers, bullet lists, and a dedicated Agent Usage Guide. Minor redundancy — the truncation note and throttle window appear twice — keeps it from a perfect score.

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

Completeness5/5

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

For a complex filtering tool paired with an almost-empty input schema and no annotations, it is exceptionally complete: full parameter semantics, validation and error contracts, freshness behavior, server-side caps, client-side filtering caveats, and three worked examples. The Returns section even documents the truncation and error shapes despite an output schema being present.

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% (filter_criteria has no type at all), so the description must fully compensate — and it does, documenting every recognized key with types, defaults, allowed values (priority 0/1/3/5), strict type rules (string "3" and float 3.0 rejected), date/timezone validation, and which keys apply to which status. The detail parameter's compact/full trade-off is also precisely explained.

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?

Opens with a specific verb and resource — "Return the tasks matching every supplied filter criterion" — making the tool's role immediately clear. The Agent Usage Guide examples and cross-references (e.g., to ticktick_get_by_id, ticktick_update_task, ticktick_sync) further set it apart from its siblings.

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

Usage Guidelines5/5

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

Explicitly tells the agent when to use alternatives: fetch full objects with ticktick_get_by_id before editing, send every field back via ticktick_update_task, and call ticktick_sync to force a refresh when local state may be stale. Three concrete invocation examples with JSON show exactly when this tool should be chosen.

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

Deploy Server

Other Tools