← Back to directory
T

Taskade MCP Server

Official
Run your Taskade workspace from any MCP-compatible AI client
Category
Collaboration #7 of 55
Stars
★ 165 Very popular
Transport
stdio (local process) · SSE
Runtime
Node.js
Credentials
API key / credential required
License
MIT
Last commit
Tools
62
62FMRS · C

Officially maintained with broad tool coverage (62 tools spanning projects, tasks, agents, templates, media, and the newer v2 Agent Chat/Webhook tools), supports both stdio and HTTP/SSE transports, and bundles a reusable OpenAPI-to-MCP code generator; note that some capabilities are gated by paid plans and the personal token carries broad permissions that should be safeguarded.

Strongest · Documentation 16/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
11/20
Maintenance
13/20
Documentation
16/20
Setup experience
14/20
Why each score
Reliability 8/20
The server.json manifest is well-formed (stdio transport, npm package, version 0.1.1) and the README enumerates 62 tools with per-category descriptions, which reads as internally consistent. However, none of the supplied material includes test files or CI workflow evidence, so there is no way to confirm the declared tool list matches actual runtime behavior or that the MCP init/tool-list handshake completes cleanly. Per the static calibration rule, reliability is capped at 12 absent verifiable execution evidence, and the score sits conservatively below that cap given the total absence of test/CI artifacts.
Security and permissions 11/20
Credential handling is reasonable: TASKADE_API_KEY is passed via env var, flagged isSecret/isRequired in the manifest, and the README explicitly warns against committing it. However, the server exposes many destructive/high-impact tools (taskDelete, deleteAgent, mediaDelete, projectComplete, agentPublicAccessEnable) with no described confirmation or safeguard mechanism, and the permission model is a single API key granting broad account-wide access with no least-privilege scoping described. Additionally, the HTTP/SSE mode passes the access token as a URL query parameter (`/sse?access_token=...`), a recognized leakage vector via logs, proxies, or referrers. The claim that the server 'talks only' to Taskade's own API is self-reported in the README and not independently verifiable from code. No red-line violation is present (no malware, covert exfiltration, or fabricated identity), but incomplete confirmation/scoping keeps this at a mid-tier score.
Maintenance 13/20
The project is MIT-licensed, not archived, includes a SECURITY.md vulnerability-reporting channel, and the README's Roadmap section shows a recent release cadence (v0.1.0 to v0.1.1) plus forward-looking plans, suggesting active development. That said, the version remains early-stage (0.1.x), and the supplied material offers no direct evidence of commit frequency, release regularity, or multi-maintainer governance, nor visibility into how the 9 open issues are being triaged, which limits the score.
Documentation 16/20
Documentation is comprehensive: concrete config snippets are provided for Claude Desktop, Cursor, Windsurf, Cline, and VS Code, plus an HTTP/SSE remote-mode section; all 62 tools are cataloged across 8 categories with short descriptions; and the README adds use-case examples, detailed Agent Recipes, and a dedicated Privacy & Security section. Gaps include the absence of per-tool parameter/input schemas (only name plus one-line description), no dedicated troubleshooting section, and no quantified rate-limit or cost disclosure (only a general note that some endpoints are gated on free plans).
Setup experience 14/20
The setup path is clear and low-friction: a single `npx -y @taskade/mcp-server` command, ready-to-paste config blocks for six mainstream clients, and documented HTTP/SSE mode with default port. The only manual step is retrieving a Personal Access Token from Taskade settings. Since the static-review rule caps setup at 15 absent real execution evidence, the score is set near that ceiling to reflect the otherwise strong documented setup flow.

Static review · not runListed 2026-08-26

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkChanges third-party account data

Best for

  • Existing Taskade users (personal or paid plans) who want an AI assistant to read and write their workspace directly
  • Project/ops managers who want to bulk-manage projects, tasks, templates, and AI agents through a conversational interface
  • Developers who want to reuse the OpenAPI codegen to quickly turn their own API specs into MCP tools

Not for

  • Free-plan users needing full functionality — some endpoints (e.g. workspace listing) are gated on free accounts
  • Anyone wanting to edit a published Genesis App's code or needing hosted OAuth access — that requires the separate Genesis App MCP at taskade.com/mcp, not this Workspace MCP
  • Anyone wanting agents to reach third-party services like Slack, Shopify, or Gmail — that's handled by Taskade's MCP Connectors, not this server

Required permissions

  • Requires a Taskade Personal Access Token (TASKADE_API_KEY) supplied as an environment variable, marked required and secret
  • The token authorizes the server to call the Taskade public API (v1 and v2) on your behalf, with read/write access to workspaces, projects, tasks, agents, media, and webhooks
  • In HTTP/SSE mode the key is passed as a query parameter (?access_token=), so consider the local network exposure

Risks and side effects

  • The token has broad read/write scope; if leaked, an attacker could create, modify, or delete your Taskade tasks, projects, media, or agents
  • Tools like agentPublicAccessEnable make an agent publicly accessible — use carefully to avoid unintentionally exposing internal content
  • Webhook subscribe/unsubscribe tools can affect other automations relying on the same webhook; misuse can break existing integrations
  • Per the README, the server only talks to Taskade's own API (api/v1 and api/v2) and does not send data to other third parties, but the API key should still be handled like any sensitive credential

Setup

Before you start

Runtime:Node.js

TASKADE_API_KEY requiredsecret Taskade personal access token, created under Taskade Settings > API at taskade.com/settings/api.
Other optional settings (1)
PORT optional Only used in --http mode to set the local HTTP server port, default 3000.

1) Go to Taskade Settings > API and create a Personal Access Token to use as TASKADE_API_KEY. 2) Add an mcpServers entry to Claude Desktop, Cursor, Windsurf, or Cline's MCP config with command npx, args ["-y", "@taskade/mcp-server"], and TASKADE_API_KEY set in env. 3) For VS Code, add a .vscode/mcp.json using the servers key; ${input:taskade_api_key} can prompt for the key on first run. 4) For remote or custom clients, run TASKADE_API_KEY=your-api-key npx @taskade/mcp-server --http to start in HTTP/SSE mode (default http://localhost:3000, configurable via PORT), then connect via SSE at /sse?access_token=your-api-key. 5) After connecting, confirm the tool list loads and grant permissions incrementally as you use write operations.

claude_desktop_config.json
{
  "mcpServers": {
    "taskade": {
      "command": "npx",
      "args": [
        "-y",
        "@taskade/mcp-server"
      ],
      "env": {
        "TASKADE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "taskade": {
      "command": "npx",
      "args": [
        "-y",
        "@taskade/mcp-server"
      ],
      "env": {
        "TASKADE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add taskade -e TASKADE_API_KEY=your-api-key-here -- npx -y @taskade/mcp-server

Run it in a terminal; replace any <…> placeholders with your own values first.

Check that it works

Check the client's tool list for the taskade server's tools (e.g., workspacesGet or meProjectsGet), or ask 'List my Taskade projects' — a successful response confirms the connection.

Troubleshooting

  1. Confirm a valid Personal Access Token was generated at Taskade Settings > API and is correctly set as TASKADE_API_KEY
  2. If npx fails to start the server, check network access to the npm registry and that the -y flag is present to skip interactive prompts
  3. Some endpoints (e.g. workspace listing) are restricted on free accounts — check your plan if you hit permission errors
  4. In --http mode, check the PORT environment variable and local port availability, and verify the SSE URL and access_token query parameter are correct
  5. For agent/knowledge tools, confirm the referenced project or media resource exists and is accessible to your account

Things to try

Once connected, you can ask your AI assistant things like:

  • List all my Taskade workspaces and projects
  • Summarize yesterday's completed tasks as a standup update and create today's priority tasks
  • Create a support AI agent trained on our docs project in the Customer Success folder and publish it publicly
  • Create a new sprint project from the Sprint Template named Sprint 2026-W15 and move the top 10 backlog tasks into it

Tools 62

workspacesGet read-only
List all workspaces
workspaceFoldersGet read-only
List folders in a workspace
workspaceCreateProject writes
Create a project in a workspace
projectGet read-only
Get project details
projectCreate writes
Create a new project
projectCopy writes
Copy a project to a folder
projectComplete writes
Mark project as completed
projectRestore writes
Restore a completed project
Show 54 more tools
projectMembersGet read-only
List project members
projectFieldsGet read-only
Get custom fields for a project
projectShareLinkGet read-only
Get the share link
projectShareLinkEnable writes
Enable the share link
projectBlocksGet read-only
Get all blocks in a project
projectTasksGet read-only
Get all tasks in a project
folderProjectsGet read-only
List projects in a folder
taskGet read-only
Get task details
taskCreate writes
Create one or more tasks
taskPut writes
Update a task
taskDelete destructive
Delete a task
taskComplete writes
Mark task as complete
taskUncomplete writes
Mark task as incomplete
taskMove writes
Move a task within a project
taskAssigneesGet read-only
Get task assignees
taskPutAssignees writes
Assign users to a task
taskDeleteAssignees writes
Remove assignees
taskGetDate read-only
Get task due date
taskPutDate writes
Set task due date
taskDeleteDate writes
Remove task due date
taskNoteGet read-only
Get task note
taskNotePut writes
Update task note
taskNoteDelete destructive
Delete task note
taskFieldsValueGet read-only
Get all field values
taskFieldValueGet read-only
Get a specific field value
taskFieldValuePut writes
Set a field value
taskFieldValueDelete destructive
Delete a field value
folderAgentGenerate writes
Generate an AI agent from a text prompt
folderCreateAgent writes
Create an agent with custom configuration
folderAgentGet read-only
List agents in a folder
agentGet read-only
Get agent details
agentUpdate writes
Update agent configuration
deleteAgent destructive
Delete an agent
agentPublicAccessEnable writes
Publish agent publicly
agentPublicGet read-only
Get public agent details
agentPublicUpdate writes
Update public agent settings
agentKnowledgeProjectCreate writes
Add a project as agent knowledge
agentKnowledgeMediaCreate writes
Add media as agent knowledge
agentKnowledgeProjectRemove writes
Remove project from knowledge
agentKnowledgeMediaRemove writes
Remove media from knowledge
agentConvosGet read-only
List agent conversations
agentConvoGet read-only
Get conversation details
publicAgentGet read-only
Get agent by public ID
folderProjectTemplatesGet read-only
List available project templates
projectFromTemplate writes
Create a project from a template
mediasGet read-only
List media files in a folder
mediaGet read-only
Get media details
mediaDelete destructive
Delete a media file
meProjectsGet read-only
List all your projects
promptAgent writes
Send a message to an AI agent and get its reply
listConversations read-only
List an agent's conversations
getConversation read-only
Get a conversation's messages
subscribeWebhook writes
Subscribe to real-time Taskade event webhooks
unsubscribeWebhook writes
Remove a webhook subscription

Use cases

View, create, and manage Taskade workspace projects and tasks via natural language in Claude/Cursor
Create, train (attach project/media knowledge to), and publicly publish Taskade AI agents through conversation
Bulk-generate projects from templates (client onboarding, sprint setup) with auto-assigned tasks and due dates
Subscribe to Taskade webhooks or hold live conversations with published agents via the v2 Agent Chat tools (beta)
Use the bundled OpenAPI-to-MCP codegen to turn any OpenAPI 3.0+ spec into MCP tools

Supported clients

Claude Desktop
Cursor
Windsurf
Cline
VS Code

Listed from the project's documentation, not tested by this site.

Overview

Taskade MCP Server (the official @taskade/mcp-server npm package) is Taskade's own Model Context Protocol server, exposing 62 tools across workspaces, projects, tasks, AI agents, knowledge bases, templates, media, and real-time webhooks to MCP clients like Claude Desktop, Cursor, Windsurf, Cline, and VS Code. It runs over stdio by default via npx, and can also run in HTTP/SSE mode (with a --http flag) for remote or custom clients. The same repository ships a separate OpenAPI-to-MCP code generator (@taskade/mcp-openapi-codegen) that can turn any OpenAPI 3.0+ spec into MCP tools, not just Taskade's own API. Note that this repo's Workspace MCP is distinct from the hosted, OAuth-protected Genesis App MCP at taskade.com/mcp described in Taskade's docs — that is a separate surface and outside the scope of this package.

Similar servers

Source revision bdaa193a624d Data synced 2026-10-12 Read the FMRS scoring method