@wizd
基本信息
TaskQueue MCP是一个革命性的企业级Agent执行引擎，核心采用BullMQ作为任务队列基础设施，实现了突破性的双循环分离架构。通过解耦用户对话循环与代理执行循环，该引擎能够执行持续数小时甚至数天的复杂工作流，同时保持完整的执行透明度与控制力。
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"taskqueue-local": {
"command": "npx",
"args": [
"-y",
"taskqueue-mcp"
]
},
"taskqueue-public": {
"url": "https://taskqueue-public.vcorp.ai/rest/your-unique-chat-id",
"stream": true
}
}
}List all projects in the system and their basic information (ID, initial prompt, task counts), optionally filtered by state (open, pending_approval, completed, all).
Read all information for a given project, by its ID, including its tasks' statuses.
Create a new project with an initial prompt and a list of tasks. This is typically the first step in any workflow.
Delete a project and all its associated tasks.
Add new tasks to an existing project.
Mark a project as complete. Can only be called when all tasks are both done and approved. This is typically the last step in a project workflow.
Use an LLM to generate a project plan and tasks from a prompt. The LLM will analyze the prompt and any attached files to create a structured project plan.
List all tasks, optionally filtered by project ID and/or state (open, pending_approval, completed, all). Tasks may include tool and rule recommendations to guide their completion.
Get details of a specific task by its ID. The task may include toolRecommendations and ruleRecommendations fields that should be used to guide task completion.
Create a new task within an existing project. You can optionally include tool and rule recommendations to guide task completion.
Modify a task's properties. Note: (1) completedDetails are required when setting status to 'done', (2) approved tasks cannot be modified, (3) status must follow valid transitions: not started → in progress → done. You can also update tool and rule recommendations to guide task completion.
Remove a task from a project.
Approve a completed task. Tasks must be marked as 'done' with completedDetails before approval. Note: This is a CLI-only operation that requires human intervention.
Get the next task to be done in a project. Returns the first non-approved task in sequence, regardless of status. The task may include toolRecommendations and ruleRecommendations fields that should be used to guide task completion.
Enterprise Task Queue Mcp Server (npm package: taskqueue-mcp) is a Model Context Protocol (MCP) server for AI task management. It helps AI assistants handle multi-step tasks in a structured way, with optional user approval checkpoints.
Configure it in an MCP client (e.g., Claude Desktop, Cursor) with the command npx -y taskqueue-mcp. Optionally set environment variables for LLM provider API keys. Use the CLI via npx taskqueue --help or npx --package=taskqueue-mcp taskqueue --help for human interaction like task approval.
It stores projects and tasks in a local JSON file. The default location is platform-specific: `~/.local/share/taskqueue-mcp/tasks.json` on Linux, `~/Library/Application Support/taskqueue-mcp/tasks.json` on macOS, and `%APPDATA%\taskqueue-mcp\tasks.json` on Windows.
Set `autoApprove` to `false` when creating a project. Then the human user must use the CLI command `npx taskqueue approve-task -- <projectId> <taskId>` to approve tasks.
Optional API keys for generating project plans using LLM providers: `OPENAI_API_KEY` (OpenAI), `GOOGLE_GENERATIVE_AI_API_KEY` (Gemini), and `DEEPSEEK_API_KEY` (Deepseek). They are not required for basic task management.
Yes, set the `TASK_MANAGER_FILE_PATH` environment variable in both the MCP client configuration and your shell before running the CLI. The same path must be used everywhere.
Tasks follow `not started → in progress → done`. From `in progress` they can go back to `not started`. From `done` they can go back to `in progress`. Approved tasks cannot be modified, and a project can only be finalized when all tasks are both done and approved.
Governed, read-only AI access to your enterprise email archive. Search years of correspondence across teams and mail systems in plain language.
Qorenext Trade screening MCP is a Model Context Protocol server that enables Claude and other AI clients to run sanctions screening and trade-screening due-diligence directly in chat.
Genchi shows engineering leaders a prediction of which projects are heading for a missed deadline and which aren't, from the people doing the work. Giving every engineer a way to be heard, this prediction comes from thei
Find wedding venues and suppliers, plan budgets and timelines, and manage wedding details, guests and events with your AI assistant. Public planning tools are free to use; private wedding tools use OAuth and changes requ
Deterministic writing filter. 43 AI patterns marked, each quoting the line that triggered it and attaching the fix. Connects as a remote MCP server at one URL. The free tier runs 3 drafts a day at 12,000 characters, no a
评论