The task manager for Cursor
Give Cursor's agent a real task queue. TaskPrio is MCP-native — Cursor pulls the top task with get_next_task, works it in your codebase, reports a result, and grabs the next. One priority order, worked top to bottom — watched live in a cockpit.
Add it to Cursor
First fetch the bundled server and verify your token (one-time — grab the token from the TaskPrio sidebar → Copy token):
PROMPTPRIO_API_TOKEN=pp_xxx bash <(curl -fsSL https://promptprio.com/install.sh)
That caches the MCP server to ~/.local/share/promptprio/mcp-server.js. Now add it to Cursor — open Cursor → Settings → MCP → Add new server (or edit ~/.cursor/mcp.json) and add promptprio under mcpServers:
{
"mcpServers": {
"promptprio": {
"command": "node",
"args": ["/Users/you/.local/share/promptprio/mcp-server.js"],
"env": {
"PROMPTPRIO_API_TOKEN": "pp_your_token",
"PROMPTPRIO_API_URL": "https://taskprio.com"
}
}
}
}
Reload Cursor — your agent now has the TaskPrio tools. Full steps in the docs.
The autopilot loop
1. get_next_task → top unblocked task + merged project context
2. (Cursor edits your codebase)
3. complete_task → short result, logged for the audit trail
4. repeat → until the queue returns no task
Each handed-out task is leased, so you can run Cursor alongside a Claude Code session on the same queue without collisions.
Why developers pair Cursor with TaskPrio
- No prompt babysitting — Cursor pulls work itself instead of you pasting the next instruction each time.
- One source of truth — the same ranked queue you see on the web board is the one Cursor reads.
- Context travels with the task — project instructions are merged into every task the agent receives.
- You set priority — capture into the Inbox, rank what matters; Cursor only works ranked tasks.
Frequently asked questions
How do I add a task queue to Cursor?
Run the one-line installer, enable the promptprio server in Cursor's MCP settings, and reload. Cursor's agent then has get_next_task and complete_task.
Does it work with Cursor's agent mode?
Yes — the agent loops get_next_task → work → complete_task on your queue, the same autopilot loop Claude Code and ChatGPT use.
Is it free?
Yes. Free includes 1 MCP session and 50 completions/day; Pro ($10/mo) gives 5 sessions and unlimited completions.
Start free — no card → Using Claude Code? →
Run it on autopilot. Paste our maintained autopilot prompt into your agent — it adds hard safety gates (money / credentials / publishing stay manual), an empty-queue stop, and auto-scopes to your project.
view ↗