{
"markdown": "\n\n# Taskade MCP Server\n\n> The **Workspace MCP** (`@taskade/mcp-server`) — run your Taskade workspace from any AI client.\n\n**Connect [Taskade](https://www.taskade.com) to any AI assistant — Claude, Cursor, Windsurf, VS Code, and more — via the [Model Context Protocol](https://modelcontextprotocol.io/).**\n\n[](https://www.npmjs.com/package/@taskade/mcp-server)\n[](https://github.com/taskade/mcp)\n[](https://github.com/taskade/mcp/blob/main/LICENSE)\n\n[](cursor://anysphere.cursor-deeplink/mcp/install?name=taskade&config=eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkB0YXNrYWRlL21jcC1zZXJ2ZXIiXSwiZW52Ijp7IlRBU0tBREVfQVBJX0tFWSI6InlvdXItYXBpLWtleS1oZXJlIn19)\n[](vscode:mcp/install?%7B%22name%22%3A%22taskade%22%2C%22command%22%3A%22npx%22%2C%22args%22%3A%5B%22-y%22%2C%22%40taskade%2Fmcp-server%22%5D%2C%22env%22%3A%7B%22TASKADE_API_KEY%22%3A%22%24%7Binput%3Ataskade_api_key%7D%22%7D%7D)\n\n**62 tools** for workspaces, projects, tasks, AI agents, agent chat, webhooks, knowledge bases, templates, media, and sharing — all from your AI client.\n\n
\n\n- [MCP Server](https://github.com/taskade/mcp/tree/main/packages/server) — Connect Taskade to Claude Desktop, Cursor, Windsurf, or any MCP client.\n- [OpenAPI Codegen](https://github.com/taskade/mcp/tree/main/packages/openapi-codegen) — Generate MCP tools from any OpenAPI spec — not just Taskade.\n\n---\n\n## Contents\n\n- [Demo](#demo)\n- [Quick Start](#quick-start)\n- [Tools (62)](#tools-62)\n- [Why Taskade MCP?](#why-taskade-mcp)\n- [Agent Recipes](#agent-recipes)\n- [Use Cases](#use-cases)\n- [OpenAPI Codegen](#openapi-codegen)\n- [What is Taskade?](#what-is-taskade)\n- [Roadmap](#roadmap)\n- [Contributing](#contributing)\n- [License](#license)\n\n---\n\n## Demo\n\nMCP-powered Taskade agent running inside Claude Desktop by Anthropic:\n\n\n\n| Build Agents via MCP | Automate Workflows | Manage Projects |\n|:---:|:---:|:---:|\n| ![\"Create](\"https://raw.githubusercontent.com/taskade/taskade/main/media/agents/agent-generator.gif\")
| ![\"Automate](\"https://raw.githubusercontent.com/taskade/taskade/main/media/automations/automation-flows.gif\")
| ![\"Manage](\"https://raw.githubusercontent.com/taskade/taskade/main/media/genesis/create-app.gif\")
|\n| Create, train, deploy AI agents from Claude/Cursor | Build multi-step automations across 100+ services | Full workspace management from your AI assistant *(paid plans; some endpoints gated on free)* |\n\n---\n\n## Quick Start\n\n> ⚡ **Prefer a hosted option?** `https://www.taskade.com/mcp` is the OAuth-protected **[Genesis App MCP](https://github.com/taskade/docs/blob/main/apis-living-system-development/genesis-app-mcp.md)** — a *separate hosted surface* from this Workspace MCP server. See its guide to learn what it does and whether it fits your use case. To read and write your workspace **content** (projects, tasks, agents) as described below, use the Workspace MCP server (this package).\n\n## Which Taskade MCP do I want?\n\nTaskade speaks MCP through three surfaces — this repo is the first one:\n\n| I want to… | Use | Where / auth |\n|---|---|---|\n| Have my AI client **read & write my projects, tasks, agents** | **Workspace MCP** — this repo (`@taskade/mcp-server`) | Local stdio · personal token · most plans |\n| **Edit my published app's code** from my IDE | **[Genesis App MCP](https://github.com/taskade/docs/blob/main/apis-living-system-development/genesis-app-mcp.md)** | Hosted · OAuth 2.0 (`https://www.taskade.com/mcp`) · Business+ |\n| Let my agents **reach Slack / Shopify / Gmail** | **[MCP Connectors](https://github.com/taskade/docs/blob/main/genesis-living-system-builder/genesis/mcp-connectors.md)** | Hosted, in-product |\n\n### 1. Get Your API Key\n\nGo to [Taskade Settings > API](https://www.taskade.com/settings/api) and create a Personal Access Token.\n\n### 2. Claude Desktop\n\nAdd to your `claude_desktop_config.json`:\n\n```json\n{\n \"mcpServers\": {\n \"taskade\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@taskade/mcp-server\"],\n \"env\": {\n \"TASKADE_API_KEY\": \"your-api-key-here\"\n }\n }\n }\n}\n```\n\n### 3. Cursor\n\nAdd to your Cursor MCP settings:\n\n```json\n{\n \"mcpServers\": {\n \"taskade\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@taskade/mcp-server\"],\n \"env\": {\n \"TASKADE_API_KEY\": \"your-api-key-here\"\n }\n }\n }\n}\n```\n\n### 4. Windsurf\n\nAdd to `~/.codeium/windsurf/mcp_config.json` (or **Settings → Cascade → MCP → Add Server**):\n\n```json\n{\n \"mcpServers\": {\n \"taskade\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@taskade/mcp-server\"],\n \"env\": {\n \"TASKADE_API_KEY\": \"your-api-key-here\"\n }\n }\n }\n}\n```\n\n### 5. Cline\n\nIn VS Code, open the Cline **MCP Servers** panel → **Configure MCP Servers** and add:\n\n```json\n{\n \"mcpServers\": {\n \"taskade\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@taskade/mcp-server\"],\n \"env\": {\n \"TASKADE_API_KEY\": \"your-api-key-here\"\n }\n }\n }\n}\n```\n\n### 6. VS Code\n\nAdd a `.vscode/mcp.json` to your workspace (VS Code uses the `servers` key; `${input:…}` prompts for your key on first run):\n\n```json\n{\n \"servers\": {\n \"taskade\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@taskade/mcp-server\"],\n \"env\": {\n \"TASKADE_API_KEY\": \"${input:taskade_api_key}\"\n }\n }\n }\n}\n```\n\n### 7. HTTP / SSE Mode (remote & custom clients)\n\n```bash\nTASKADE_API_KEY=your-api-key npx @taskade/mcp-server --http\n```\n\nThe server starts at `http://localhost:3000` (configure with `PORT` env var). Connect via SSE at `http://localhost:3000/sse?access_token=your-api-key`.\n\n---\n\n## Tools (62)\n\n### Workspaces\n\n| Tool | Description |\n|------|-------------|\n| `workspacesGet` | List all workspaces |\n| `workspaceFoldersGet` | List folders in a workspace |\n| `workspaceCreateProject` | Create a project in a workspace |\n\n### Projects\n\n| Tool | Description |\n|------|-------------|\n| `projectGet` | Get project details |\n| `projectCreate` | Create a new project |\n| `projectCopy` | Copy a project to a folder |\n| `projectComplete` | Mark project as completed |\n| `projectRestore` | Restore a completed project |\n| `projectMembersGet` | List project members |\n| `projectFieldsGet` | Get custom fields for a project |\n| `projectShareLinkGet` | Get the share link |\n| `projectShareLinkEnable` | Enable the share link |\n| `projectBlocksGet` | Get all blocks in a project |\n| `projectTasksGet` | Get all tasks in a project |\n| `folderProjectsGet` | List projects in a folder |\n\n### Tasks\n\n| Tool | Description |\n|------|-------------|\n| `taskGet` | Get task details |\n| `taskCreate` | Create one or more tasks |\n| `taskPut` | Update a task |\n| `taskDelete` | Delete a task |\n| `taskComplete` | Mark task as complete |\n| `taskUncomplete` | Mark task as incomplete |\n| `taskMove` | Move a task within a project |\n| `taskAssigneesGet` | Get task assignees |\n| `taskPutAssignees` | Assign users to a task |\n| `taskDeleteAssignees` | Remove assignees |\n| `taskGetDate` | Get task due date |\n| `taskPutDate` | Set task due date |\n| `taskDeleteDate` | Remove task due date |\n| `taskNoteGet` | Get task note |\n| `taskNotePut` | Update task note |\n| `taskNoteDelete` | Delete task note |\n| `taskFieldsValueGet` | Get all field values |\n| `taskFieldValueGet` | Get a specific field value |\n| `taskFieldValuePut` | Set a field value |\n| `taskFieldValueDelete` | Delete a field value |\n\n### AI Agents\n\nCreate, manage, and publish autonomous AI agents with custom knowledge and tools.\n\n| Tool | Description |\n|------|-------------|\n| `folderAgentGenerate` | Generate an AI agent from a text prompt |\n| `folderCreateAgent` | Create an agent with custom configuration |\n| `folderAgentGet` | List agents in a folder |\n| `agentGet` | Get agent details |\n| `agentUpdate` | Update agent configuration |\n| `deleteAgent` | Delete an agent |\n| `agentPublicAccessEnable` | Publish agent publicly |\n| `agentPublicGet` | Get public agent details |\n| `agentPublicUpdate` | Update public agent settings |\n| `agentKnowledgeProjectCreate` | Add a project as agent knowledge |\n| `agentKnowledgeMediaCreate` | Add media as agent knowledge |\n| `agentKnowledgeProjectRemove` | Remove project from knowledge |\n| `agentKnowledgeMediaRemove` | Remove media from knowledge |\n| `agentConvosGet` | List agent conversations |\n| `agentConvoGet` | Get conversation details |\n| `publicAgentGet` | Get agent by public ID |\n\n### Templates\n\n| Tool | Description |\n|------|-------------|\n| `folderProjectTemplatesGet` | List available project templates |\n| `projectFromTemplate` | Create a project from a template |\n\n### Media\n\n| Tool | Description |\n|------|-------------|\n| `mediasGet` | List media files in a folder |\n| `mediaGet` | Get media details |\n| `mediaDelete` | Delete a media file |\n\n### Personal\n\n| Tool | Description |\n|------|-------------|\n| `meProjectsGet` | List all your projects |\n\n### Agent Chat & Webhooks (API v2, beta)\n\nCapabilities the v1 API doesn't have: hold a live conversation with an AI agent, and receive real-time events.\n\n| Tool | Description |\n|------|-------------|\n| `promptAgent` | Send a message to an AI agent and get its reply |\n| `listConversations` | List an agent's conversations |\n| `getConversation` | Get a conversation's messages |\n| `subscribeWebhook` | Subscribe to real-time Taskade event webhooks |\n| `unsubscribeWebhook` | Remove a webhook subscription |\n\n---\n\n## Why Taskade MCP?\n\nTaskade MCP gives your AI assistant **access to your workspace** — projects, tasks, agents, knowledge bases, templates, and automations (on paid plans; some endpoints such as workspace listing are gated for free accounts). Instead of writing API code, describe what you need in natural language.\n\n```\n┌──────────────────────────────────────────────────────────┐\n│ HOW TASKADE MCP WORKS │\n└──────────────────────────────────────────────────────────┘\n\n You (in Claude/Cursor/Windsurf):\n \"Create a support agent trained on our docs project\"\n ↓\n ┌──────────────────────────────────┐\n │ Taskade MCP Server │\n │ (62 tools, 8 categories) │\n └──────────────────────────────────┘\n ↓ ↓ ↓\n folderCreateAgent agentKnowledge agentPublicAccess\n (creates agent) (attaches docs) (publishes it)\n ↓\n Result: Live AI chatbot trained on your data\n```\n\n| What You Say | What Happens |\n|-------------|-------------|\n| \"Show me all overdue tasks\" | Reads projects → filters tasks → formats report |\n| \"Create a support agent trained on our docs\" | Creates agent → attaches knowledge → publishes |\n| \"Set up next week's sprint from our template\" | Copies template → populates tasks → assigns team |\n| \"Summarize yesterday's completed work\" | Reads task history → generates standup summary |\n\n### Why MCP Instead of REST API?\n\n| | REST API | MCP Server |\n|-|----------|-----------|\n| **Setup** | Write HTTP client, handle auth, parse JSON | `npx @taskade/mcp-server` — one command |\n| **Interface** | Code against endpoints | Natural language — describe what you need |\n| **Chaining** | Manual orchestration | AI chains tools intelligently |\n| **Error handling** | Write error handling code | AI interprets errors and retries |\n\n### Why Taskade MCP Over Other MCP Servers?\n\nTaskade is the only MCP server that includes **AI agent management** (create, train, deploy agents), **knowledge base training** (attach docs, projects, media), and **OpenAPI codegen** (generate MCP tools from any API spec). 62 tools across 8 categories.\n\n---\n\n## Agent Recipes\n\n\nRecipe 1: Daily Standup Summarizer
\n
\n\n**Problem:** Team standup notes scattered across Slack, email, and docs.\n\n```\nYou: \"Check my Taskade workspace for yesterday's completed tasks,\n summarize them as a standup update, then create today's\n priority tasks based on what's still open.\"\n\nClaude uses:\n 1. meProjectsGet → find your active projects\n 2. projectTasksGet → pull tasks from each project\n 3. taskCreate → create today's priority tasks\n\nResult: A formatted standup summary + fresh task list in Taskade.\n```\n \n\n\nRecipe 2: Knowledge Base Agent Builder
\n
\n\n**Problem:** You have docs scattered across files and need an AI agent trained on them.\n\n```\nYou: \"Create an AI support agent called 'Help Bot' in my\n Customer Success folder, train it on our Documentation\n project, and publish it publicly.\"\n\nClaude uses:\n 1. workspaceFoldersGet → find \"Customer Success\" folder\n 2. folderCreateAgent → create \"Help Bot\" with support instructions\n 3. agentKnowledgeProjectCreate → attach Documentation project\n 4. agentPublicAccessEnable → publish with shareable link\n\nResult: Live AI chatbot trained on your docs, ready to embed.\n```\n \n\n\nRecipe 3: Sprint Planning Automation
\n
\n\n**Problem:** Creating sprint projects manually every two weeks.\n\n```\nYou: \"Create a new sprint project from our 'Sprint Template',\n name it 'Sprint 2026-W15', move the top 10 backlog\n items into it, and assign them to the engineering team.\"\n\nClaude uses:\n 1. folderProjectTemplatesGet → find \"Sprint Template\"\n 2. projectFromTemplate → create \"Sprint 2026-W15\"\n 3. projectTasksGet → get backlog items\n 4. taskMove → move top 10 tasks to new sprint\n 5. taskPutAssignees → assign engineering team members\n\nResult: Sprint ready to go, fully populated and assigned.\n```\n \n\n\nRecipe 4: Competitive Intelligence Agent
\n
\n\n**Problem:** You need an agent that monitors competitors and reports findings.\n\n```\nYou: \"Generate an AI agent that researches our top 5 competitors.\n Train it on our Competitive Analysis project.\n Name it 'Market Intel Agent'.\"\n\nClaude uses:\n 1. folderAgentGenerate → generate agent from research prompt\n 2. agentKnowledgeProjectCreate → connect Competitive Analysis project\n 3. agentUpdate → refine instructions for research cadence\n\nResult: AI research agent with domain knowledge, ready for briefings.\n```\n \n\n\nRecipe 5: Client Onboarding Pipeline
\n
\n\n**Problem:** Each new client needs the same project structure, tasks, and materials.\n\n```\nYou: \"For our new client Acme Corp: create a project from\n the 'Client Onboarding' template, add tasks for\n kickoff meeting and SOW review, set due dates for\n the next 2 weeks, then share the project link.\"\n\nClaude uses:\n 1. projectFromTemplate → create \"Acme Corp Onboarding\"\n 2. taskCreate → add kickoff, requirements, SOW tasks\n 3. taskPutDate → set dates across next 2 weeks\n 4. projectShareLinkEnable → generate share link\n\nResult: Client onboarding project live and shareable in 30 seconds.\n```\n \n\n---\n\n## Use Cases\n\n### Project Management with AI\n\nAsk your AI assistant to manage your Taskade workspace:\n\n- \"Show me all my projects and their status\"\n- \"Create a new project called Q1 Planning with tasks for each team\"\n- \"Move all overdue tasks to the Backlog project\"\n- \"Set due dates for all tasks in the Sprint project\"\n\n### AI Agent Creation\n\nBuild and deploy AI agents directly from your editor:\n\n- \"Create an AI agent called Customer Support Bot with knowledge from our docs project\"\n- \"Generate an agent for code review using this prompt: ...\"\n- \"Publish my agent publicly and give me the share link\"\n- \"Add our API documentation project as knowledge to the agent\"\n\n### Template Workflows\n\nAutomate project creation from templates:\n\n- \"List all templates in my workspace\"\n- \"Create 5 new client onboarding projects from the Client Template\"\n- \"Copy the Sprint Retrospective project for this week\"\n\n---\n\n## OpenAPI Codegen\n\nUse our generator to build MCP tools from any OpenAPI spec — not just Taskade.\n\n```bash\nnpm install --save-dev @taskade/mcp-openapi-codegen @readme/openapi-parser\n```\n\n```ts\nimport { dereference } from '@readme/openapi-parser';\nimport { codegen } from '@taskade/mcp-openapi-codegen';\n\nconst document = await dereference('your-api-spec.yaml');\n\nawait codegen({\n path: 'src/tools.generated.ts',\n document,\n});\n```\n\nWorks with any OpenAPI 3.0+ spec. Generate MCP tools for your own APIs in minutes.\n\n---\n\n## What is Taskade?\n\n[Taskade](https://www.taskade.com) ([Y Combinator S19](https://www.ycombinator.com/companies/taskade)) is the AI-native workspace for building apps, deploying agents, and automating workflows — from a single prompt. **150,000+ apps generated. Trusted by 3M, Nike, Tesla, Netflix, Airbnb, Disney, Adobe.** Rated 4.8/5 across 9,300+ reviews.\n\n- **Genesis Apps** — Build complete apps from prompts. Dashboards, CRMs, portals, forms — deployed instantly. [Try it →](https://www.taskade.com/create)\n- **AI Agents** — Custom agents with 22+ tools, persistent memory, multi-agent teams, public embedding\n- **Automations** — No-code workflow automation with 100+ integrations, branching, looping, filtering\n- **Real-time Collaboration** — Multiplayer workspace with chat, video, 7 project views\n- **Templates** — 700+ templates for project management, engineering, marketing, and more\n- **API & MCP** — REST API v2, this MCP Server, Agent API, webhooks, OAuth 2.0\n\n**Links:**\n- App: [taskade.com](https://www.taskade.com)\n- Create: [taskade.com/create](https://www.taskade.com/create)\n- Agents: [taskade.com/agents](https://www.taskade.com/agents)\n- Templates: [taskade.com/templates](https://www.taskade.com/templates)\n- Community: [taskade.com/community](https://www.taskade.com/community)\n- Developer Docs: [docs.taskade.com](https://docs.taskade.com)\n- **Workspace MCP guide:** https://github.com/taskade/docs/blob/main/apis-living-system-development/workspace-mcp.md\n- **Genesis App MCP guide** (a different, hosted surface): https://github.com/taskade/docs/blob/main/apis-living-system-development/genesis-app-mcp.md\n- Blog: [taskade.com/blog](https://www.taskade.com/blog)\n\n---\n\n## Roadmap\n\nSee [open issues](https://github.com/taskade/mcp/issues) for planned features and improvements.\n\n**Recently shipped (v0.1.0):** Agent chat (`promptAgent`) and webhook triggers, live now via the [Taskade API v2 tool layer](#agent-chat--webhooks-api-v2-beta).\n\n- **Automation & Flow Tools** — Create, enable, and manage workflow automations via MCP\n- **`agent.js`** — Open-source autonomous agent toolkit (coming soon)\n- **TaskOS** — Agent platform at [docs.taskade.com](https://docs.taskade.com)\n\n---\n\n## Privacy & Security\n\nYour Taskade API token authorizes the MCP server to call the Taskade public API on your behalf. The server talks **only** to the Taskade public API — `https://www.taskade.com/api/v1` and `https://www.taskade.com/api/v2` — and does not send your data to other third-party services.\n\n- **Privacy** — see the [Taskade Privacy Policy](https://www.taskade.com/privacy).\n- **Security & vulnerability reporting** — see [SECURITY.md](./SECURITY.md).\n- Store your token in an environment variable; never commit it.\n\n---\n\n## Contributing\n\nHelp us improve MCP tools, OpenAPI workflows, and agent capabilities.\n\n- [Issues](https://github.com/taskade/mcp/issues) — Report bugs or request features\n- [Pull Requests](https://github.com/taskade/mcp/pulls) — Contributions welcome\n- [Community](https://www.taskade.com/community) — Join the Taskade community\n- [Contact](mailto:hello@taskade.com) — hello@taskade.com\n\n---\n\n## More from Taskade\n\n- [Taskade Docs](https://docs.taskade.com) ([source](https://github.com/taskade/docs)) — developer documentation and API reference\n- [Integration Kit](https://github.com/taskade/integrations) — open-source Zapier/n8n actions & triggers\n- [Taskade](https://github.com/taskade/taskade) — platform home, including Genesis, the AI app builder\n\n---\n\n## License\n\nMIT\n",
"bytes": 20505,
"sha": "04c8f70a074435c8e2c00827d6a8edbb4445026d2c3924489c2e1d77dbb02a7e",
"repo_slug": "taskade/mcp",
"fonte": "repo",
"truncated": false,
"api": "https://agentalog.com/api/listings/mcp_io_github_taskade_mcp_server_b4314b26/readme"
}