# Taskhold (remote · taskhold.com) Shared task list for Claude Code, Cursor and Codex: live progress per step, questions you answer. - Trust score: 83/100 (high trust) - Change this week: +5 - Registry status: active - Liveness: live - Owner verified: no - Last scored: 2026-11-05 ## Components - remote · `taskhold.com`: 83/100 (this document), [markdown](https://verifymcp.io/servers/com-taskhold-taskhold/api-mcp.md), [page](https://verifymcp.io/servers/com-taskhold-taskhold/api-mcp) ## Channel facts - Endpoint: `https://taskhold.com/api/mcp` - Transports: `streamable-http` - Auth: `required` - Version: `1.0.4` ## Trust breakdown How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog) Scored 2026-11-05. - **Endpoint Security**: 97/100 - The endpoint's TLS certificate is valid, in date, and uses a strong key. - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token. - HTTPS is enforced; there's no plaintext access path. - The HSTS (Strict-Transport-Security) header is present. - DNSSEC check failed: this domain isn't protected by DNSSEC. - The authorisation server supports Client ID Metadata Documents, the current MCP client-registration mechanism. - **Transport & Reachability**: 100/100 - Verified streamable-http transport via a live MCP handshake. - **Schema Quality & AI Usability**: 66/100 - AI-judged instruction clarity (excellent). - Context-footprint check failed: tool/resource definitions use about 4247 tokens (~193/item across 22 items; 22 tools + 0 resources), over budget; trim descriptions and params. - Usage-examples check failed: none of the tools include examples. - **Stability & Change Management**: 33/100 - Stability observed for 10 of 30 days with no destabilising changes; credit accrues until the full window elapses. - **Tool Coverage**: 100/100 - 100% of tools have a non-trivial description (not blank, and not just the tool's name). - 100% of tool parameters carry a description. - **Tool Safety**: 100/100 - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured. - We read all 22 captured tool definition(s), and no name or description among them implies an irreversible operation. - An AI judge read all 23 captured unit(s) of tool text and found none that tries to manipulate the model reading it. - **Capabilities**: 100/100 - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28. ## Install ### How do I install the Taskhold MCP server? Taskhold is a hosted endpoint at https://taskhold.com/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation. ### Claude ```bash claude mcp add --transport http com-taskhold-taskhold 'https://taskhold.com/api/mcp' ``` ### Cursor ```json { "mcpServers": { "com-taskhold-taskhold": { "url": "https://taskhold.com/api/mcp" } } } ``` ### VS Code ```json { "servers": { "com-taskhold-taskhold": { "type": "http", "url": "https://taskhold.com/api/mcp" } } } ``` ### Codex ```toml [mcp_servers.com-taskhold-taskhold] url = "https://taskhold.com/api/mcp" ``` ### opencode ```json { "$schema": "https://opencode.ai/config.json", "mcp": { "com-taskhold-taskhold": { "type": "remote", "url": "https://taskhold.com/api/mcp", "enabled": true } } } ``` ### OpenClaw ```bash openclaw mcp add com-taskhold-taskhold --url 'https://taskhold.com/api/mcp' --transport streamable-http ``` ### Hermes ```yaml mcp_servers: com-taskhold-taskhold: url: "https://taskhold.com/api/mcp" ``` ### Netclaw ```json { "McpServers": { "com-taskhold-taskhold": { "Transport": "http", "Url": "https://taskhold.com/api/mcp" } } } ``` ### Vellum ```bash assistant mcp add com-taskhold-taskhold -t streamable-http -u 'https://taskhold.com/api/mcp' ``` ### Other ```json { "mcpServers": { "com-taskhold-taskhold": { "type": "http", "url": "https://taskhold.com/api/mcp" } } } ``` The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs. ## Changelog Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims. ### 2026-11-05 (score 83, 0) - [functional] Server version: 1.5.5 → 1.5.6 ### 2026-11-04 (score 83, +1) No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-11-02 (score 82, +1) No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-11-01 (score 81, 0) - [functional] Server version: 1.5.4 → 1.5.5 ### 2026-10-31 (score 81, +3) - [security improvement] Authorization: fail → pass - [security] First check of Authorization: pass - [security] New tool “taskhold_update_project”, which the server declares destructive - [security] Tool “taskhold_add_note” rewrote its description, which is the text the model reads - [security] Tool “taskhold_begin” rewrote its description, which is the text the model reads - [security] Tool “taskhold_block” rewrote its description, which is the text the model reads - [security] Tool “taskhold_complete” rewrote its description, which is the text the model reads - [security] Tool “taskhold_create” rewrote its description, which is the text the model reads - [security] Tool “taskhold_create_from_todo” rewrote its description, which is the text the model reads - [security] Tool “taskhold_create_project” rewrote its description, which is the text the model reads - [security] Tool “taskhold_get” rewrote its description, which is the text the model reads - [security] Tool “taskhold_get_current” rewrote its description, which is the text the model reads - [security] Tool “taskhold_get_summary” rewrote its description, which is the text the model reads - [security] Tool “taskhold_list” rewrote its description, which is the text the model reads - [security] Tool “taskhold_list_labels” rewrote its description, which is the text the model reads - [security] Tool “taskhold_list_projects” rewrote its description, which is the text the model reads - [security] Tool “taskhold_plan” rewrote its description, which is the text the model reads - [security] Tool “taskhold_progress” rewrote its description, which is the text the model reads - [security] Tool “taskhold_search” rewrote its description, which is the text the model reads - [security] Tool “taskhold_set_priority” rewrote its description, which is the text the model reads - [security] Tool “taskhold_start” rewrote its description, which is the text the model reads - [security] Tool “taskhold_unblock” rewrote its description, which is the text the model reads - [security] Tool “taskhold_update” rewrote its description, which is the text the model reads - [security] Tool “taskhold_updates” rewrote its description, which is the text the model reads - [security] Tool “taskhold_plan” is now declared destructive - [security] Tool “taskhold_update” is now declared destructive - [functional regression] Schema quality: 126 → 193 - [functional] Schema quality: good → excellent - [functional] Server version: 1.2.0 → 1.5.4 - [cosmetic] “taskhold_block” added an optional parameter “if_no_answer” - [cosmetic] “taskhold_block” added an optional parameter “options” - [cosmetic] “taskhold_block” added an optional parameter “recommended” - [cosmetic] “taskhold_complete” added an optional parameter “finish_task” - [cosmetic] “taskhold_complete” added an optional parameter “needs_review” - [cosmetic] “taskhold_complete” added an optional parameter “outcome” - [cosmetic] “taskhold_create_from_todo” added an optional parameter “labels” - [cosmetic] “taskhold_create_from_todo” added an optional parameter “priority” - [cosmetic] “taskhold_search” added an optional parameter “project_id” - [cosmetic] “taskhold_begin” reworded the description of “labels” - [cosmetic] “taskhold_complete” reworded the description of “summary” - [cosmetic] Tool “taskhold_add_note” changed its title: Add a note - [cosmetic] Tool “taskhold_begin” changed its title: Begin a task - [cosmetic] Tool “taskhold_block” changed its title: Block a task with a question - [cosmetic] Tool “taskhold_complete” changed its title: Complete a task - [cosmetic] Tool “taskhold_create” changed its title: Create a task - [cosmetic] Tool “taskhold_create_from_todo” changed its title: Create a task from a TODO - [cosmetic] Tool “taskhold_create_project” changed its title: Create a project - [cosmetic] Tool “taskhold_get” changed its title: Get a task - [cosmetic] Tool “taskhold_get_current” changed its title: Get current tasks - [cosmetic] Tool “taskhold_get_summary” changed its title: Get a summary - [cosmetic] Tool “taskhold_list” changed its title: List tasks - [cosmetic] Tool “taskhold_list_labels” changed its title: List labels - [cosmetic] Tool “taskhold_list_projects” changed its title: List projects - [cosmetic] Tool “taskhold_plan” changed its title: Plan task steps - [cosmetic] Tool “taskhold_progress” changed its title: Report progress - [cosmetic] Tool “taskhold_search” changed its title: Search tasks - [cosmetic] Tool “taskhold_set_priority” changed its title: Set priority - [cosmetic] Tool “taskhold_start” changed its title: Start a task - [cosmetic] Tool “taskhold_unblock” changed its title: Unblock a task - [cosmetic] Tool “taskhold_update” changed its title: Edit a task - [cosmetic] Tool “taskhold_updates” changed its title: Get the person's updates ### 2026-10-30 (score 78, 0) - [security] The server rewrote its instructions, which are the text every model session reads - [security] Tool “taskhold_start” rewrote its description, which is the text the model reads - [security] Tool “taskhold_add_note” rewrote its description, which is the text the model reads - [security] Tool “taskhold_block” rewrote its description, which is the text the model reads - [security] Tool “taskhold_complete” rewrote its description, which is the text the model reads - [security] Tool “taskhold_create” rewrote its description, which is the text the model reads - [security] Tool “taskhold_get” rewrote its description, which is the text the model reads - [security] Tool “taskhold_get_current” rewrote its description, which is the text the model reads - [security] Tool “taskhold_list” rewrote its description, which is the text the model reads - [security] Tool “taskhold_list_projects” rewrote its description, which is the text the model reads - [security] Tool “taskhold_plan” rewrote its description, which is the text the model reads - [security] Tool “taskhold_progress” rewrote its description, which is the text the model reads - [security] Tool “taskhold_search” rewrote its description, which is the text the model reads - [security] Tool “taskhold_unblock” rewrote its description, which is the text the model reads - [security] Tool “taskhold_updates” rewrote its description, which is the text the model reads - [functional regression] Schema quality: pass → fail - [functional] New tool “taskhold_begin” - [functional] New tool “taskhold_create_project” - [cosmetic] “taskhold_complete” added an optional parameter “start_next” - [cosmetic] “taskhold_create” added an optional parameter “project” - [cosmetic] “taskhold_create_from_todo” added an optional parameter “project” - [cosmetic] “taskhold_get_current” added an optional parameter “project” - [cosmetic] “taskhold_list” added an optional parameter “project” - [cosmetic] “taskhold_search” added an optional parameter “project” - [cosmetic] “taskhold_update” added an optional parameter “project” - [cosmetic] “taskhold_updates” added an optional parameter “project” - [cosmetic] “taskhold_updates” added an optional parameter “task_id” - [cosmetic] “taskhold_updates” added an optional parameter “wait” - [cosmetic] “taskhold_complete” reworded the description of “summary” - [cosmetic] “taskhold_create” reworded the description of “project_id” - [cosmetic] “taskhold_list” reworded the description of “project_id” - [cosmetic] “taskhold_progress” reworded the description of “task_id” - [cosmetic] “taskhold_start” reworded the description of “agent” - [cosmetic] “taskhold_update” reworded the description of “project_id” - [cosmetic] “taskhold_updates” reworded the description of “include_agent” - [cosmetic] “taskhold_updates” reworded the description of “since” ### 2026-10-28 (score 78, +1) No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-10-27 (score 77, 0) - [security] The server rewrote its instructions, which are the text every model session reads - [security] Tool “taskhold_block” rewrote its description, which is the text the model reads - [functional improvement] Stability: unverified → 0.03 - [cosmetic] “taskhold_block” reworded the description of “reason” ## MCP tools (22) ### `taskhold_begin` (~394 tokens) Begin a task Start work in one call: files a new task (or takes an existing one with task_id), plans its steps, marks it in progress and starts the first step. Use this whenever the person asks for something: it puts the request on their board with a checklist and a progress bar. Returns the task id, the numbered steps with ids, the step now running and a cursor for taskhold_updates. Every event of the call (created, planned, started) carries the agent label. On a task already under way it appends the steps. Fails without title or task_id, on an unknown task_id (get one from taskhold_list) or on a step (begin its parent, or taskhold_start the step). Input parameters: - `agent` (string): Who is working, shown to the person: " in ", e.g. "Claude Code in api-server". Outside a repo just "Claude Code". - `description` (string): Details (markdown), e.g. the person's request in full - `labels` (array): Labels for a new task, e.g. ["bug", "refactor"] - `priority` (string): Priority of a new task (default medium) - `project` (string): Project name, e.g. the repo folder ("api-server"). Matched case-insensitively; created on first use. Use this instead of project_id. - `start_first_step` (boolean): Also start the first step (default true) - `steps` (array): The plan, in order, e.g. ["Read the code", "Write the migration", "Add tests"]. Omit for a one-step job. - `task_id` (string): Begin an existing task instead of filing a new one (from taskhold_list). - `title` (string): Title of the new task (the person's request, short). Omit when passing task_id. ### `taskhold_list` (~164 tokens) List tasks List top-level tasks, one line each, grouped by status. Pass project to see one repo's work. Each line has status, title, steps done, progress and id; steps aren't listed (taskhold_get shows them). An unknown project name returns no tasks and is not created. Input parameters: - `limit` (integer): Max tasks to return (default 20) - `priority` (string): Filter by priority - `project` (string): Project name, e.g. the repo folder ("api-server"). Matched case-insensitively (or a project id). An unknown name matches nothing. - `project_id` (string): Filter by project id (project, a name, is easier) - `status` (string): Filter by status. Default: all. ### `taskhold_get` (~130 tokens) Get a task Get one task with its full description, steps (subtasks), progress, labels and the latest notes. Returns status, priority, project, the numbered steps with ids (each with its description and, once done, its summary), up to 10 latest notes, the completion summary and what changed (files, checks, links, risks, how to verify) when it was reported. An unknown task_id fails with "Task not found"; get a valid id from taskhold_list (status "all") or taskhold_search. Input parameters: - `task_id` (string, required): The ID of the task to retrieve ### `taskhold_get_current` (~147 tokens) Get current tasks Get the task in progress (in this project when you pass project), with its steps and the next one to do. Call this at the start of a session. Returns the most recently active in-progress task, its numbered steps with ids (with descriptions and the summaries of done steps), the next step's id and the latest 3 notes (taskhold_get shows 10); with nothing in progress (or an unknown project) it says so and names taskhold_begin and taskhold_list. Input parameters: - `project` (string): Project name, e.g. the repo folder ("api-server"). Matched case-insensitively (or a project id). An unknown name matches nothing. ### `taskhold_search` (~135 tokens) Search tasks Search tasks and steps by text in the title or description (case-insensitive, up to 100 tasks scanned). Returns matching lines grouped by status, with ids (a step names the task it belongs to), or a line saying nothing matched. Input parameters: - `project` (string): Project name, e.g. the repo folder ("api-server"). Matched case-insensitively (or a project id). An unknown name matches nothing. - `project_id` (string): Filter by project id (project, a name, is easier) - `query` (string, required): Search text - `status` (string): Filter by status ### `taskhold_create` (~244 tokens) Create a task File work for later: a bug, a refactor, a follow-up you aren't doing now (it lands in To Do). To file and start work at once, use taskhold_begin. Pass parent_task_id to add one step to an existing task (taskhold_plan for several). Returns the new id, status (todo) and priority. Fails if parent_task_id isn't found or is itself a step (steps don't nest), if project_id doesn't exist (pass project by name instead), or at the account's task ceiling. Input parameters: - `description` (string): Details (markdown) - `labels` (array): Labels, e.g. ["bug", "refactor"] - `parent_task_id` (string): Make it a step (subtask) of this top-level task - `priority` (string): Priority (default medium) - `project` (string): Project name, e.g. the repo folder ("api-server"). Matched case-insensitively; created on first use. Use this instead of project_id. - `project_id` (string): Project id (project, a name, is easier) - `title` (string, required): Task title ### `taskhold_plan` (~183 tokens) Plan task steps Add ordered steps (subtasks) to a task so the person sees a checklist; the progress bar fills as steps finish. taskhold_begin already does this for new work. For an existing task: taskhold_start it, then taskhold_plan. Steps don't nest. Returns the task's whole step list with ids. Fails on an unknown task_id, on a step (plan its parent) or past 200 steps per task. Planning a done task adds todo steps but leaves it done; taskhold_start it to reopen. Input parameters: - `replace` (boolean): Replace steps that haven't started yet (done and in-progress steps stay). Default false: append. - `steps` (array, required): Steps in order, e.g. ["Write the migration", "Update the API", "Add tests"] - `task_id` (string, required): The ID of the task to plan ### `taskhold_progress` (~214 tokens) Report progress Move the person's progress bar during a long step or a task without steps: percent, or current and total (e.g. 12 of 40 files), plus a short message about what's happening now. On a task with steps, the steps drive the bar: numbers you send for the task go to its running step. Call at milestones, every few minutes at most. Returns the task's progress bar and, on a task with steps, which step your numbers went to. Fails without percent, total or message, on an unknown task_id, or past the per-minute limit (the error says when to retry). Input parameters: - `current` (integer): Items done so far, with total - `message` (string): What is happening now, e.g. "Running the test suite" - `percent` (number): 0-100 - `task_id` (string, required): The ID of the task you are reporting on (a step, or a task without steps) - `total` (integer): Total items ### `taskhold_create_from_todo` (~209 tokens) Create a task from a TODO Create a task from a TODO comment in code, keeping the file path and line number (label todo-from-code plus any labels you pass; priority as in taskhold_create; use taskhold_create for anything else). Returns the new task with its id and code location. Fails at the account's task ceiling or on an invalid project name. Input parameters: - `description` (string): Extra context from the code - `file_path` (string, required): Path of the file containing the TODO - `labels` (array): More labels besides todo-from-code, e.g. ["bug"] - `line_number` (integer, required): Line number of the TODO - `priority` (string): Priority (default medium; low unless it's a real bug) - `project` (string): Project name, e.g. the repo folder ("api-server"). Matched case-insensitively; created on first use. Use this instead of project_id. - `title` (string, required): Task title from the TODO ### `taskhold_start` (~141 tokens) Start a task Mark a task or step in progress when you begin it. Starting a step also starts its task. For new work, taskhold_begin files, plans and starts in one call. Returns the id, status and the next step's id when it has steps. Safe to repeat; it also resumes a blocked or done task. An unknown task_id fails with "Task not found". Input parameters: - `agent` (string): Who is working, shown to the person: " in ", e.g. "Claude Code in api-server". Outside a repo just "Claude Code". - `task_id` (string, required): The ID of the task to start ### `taskhold_complete` (~423 tokens) Complete a task Mark a task or step done, with a summary of what you did and, optionally, a structured outcome: files changed (with +/- line counts), checks run (pass/fail/skip), links (PR, commit, branch, deploy, doc), follow-up task ids, risks and how the person can verify it. The person sees it as a What changed card; a step's outcome rolls up to its task. When the person should look at the work, pass needs_review: true with outcome.verify (how you checked it): the task lands in their Ready for review queue. Completing a step tells you the next one; start_next: true also starts it, and finish_task: true on the last open step closes the task too (with the same summary) in one call. For a step it returns steps done, the task's bar and the next step's id (or, after the last step, a reminder to complete the task). Completing a task closes its open steps. Repeating it on a task that is already done changes nothing and keeps the first summary (add more with taskhold_add_note); an unknown task_id fails with "Task not found", an invalid outcome fails without changing anything. Input parameters: - `finish_task` (boolean): When completing the last open step, also complete its task with this summary (default false) - `needs_review` (boolean): Ask the person to review the work: the task goes to their Ready for review queue (it still counts as done). Needs outcome.verify, a short "how I checked it" line. They approve it or request changes;… - `outcome` (object): What changed, shown to the person as a What changed card (20 KB max). Every field is optional. - `start_next` (boolean): When completing a step, start the next step too (default false) - `summary` (string): What was done (and how you checked it). Markdown. - `task_id` (string, required): The ID of the task to complete ### `taskhold_block` (~371 tokens) Block a task with a question Mark a task (or step) blocked when you can't continue without the person (a decision, missing info, access). Returns the question and the exact taskhold_updates call to wait with. Write the reason as one question they can answer from the board in a line, e.g. "Decide: drop the legacy column or keep it nullable?". When the answer is a choice, pass 2-4 short options (and recommended, the index of the one you'd pick, 0 = first): the person answers with one tap and the reply reads Reply to "": chose "