What's the difference between scheduled and on-demand tasks?
Scheduled tasks include a cron expression and run automatically on that schedule. On-demand tasks omit the schedule and sit idle until triggered via the `run_task` tool.
@jolks
Basic information
MCP server for scheduling and running AI prompts, HTTP/webhook requests, and shell commands
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"mcp-cron": {
"command": "npx",
"args": [
"-y",
"mcp-cron"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
Fetch the live tool list by running this server in a temporary sandbox using the button above.
MCP Cron is a Model Context Protocol server for scheduling and managing tasks through a standardized API. It supports shell commands, AI-powered tasks, and HTTP webhook tasks, with task persistence across restarts via SQLite. It's intended for developers and AI agents that need to schedule recurring or on-demand work.
Install via npx -y mcp-cron and configure it in your MCP client's config file (e.g., ~/.cursor/mcp.json or Claude Desktop's config). The server supports two transport modes: HTTP (Streamable HTTP, default) and stdio. Key command-line arguments include --transport, --ai-provider, --ai-model, --db-path, and --prevent-sleep.
run_taskScheduled tasks include a cron expression and run automatically on that schedule. On-demand tasks omit the schedule and sit idle until triggered via the `run_task` tool.
It supports HTTP (Streamable HTTP, the default) and stdio. HTTP is for browser/network clients; stdio is for direct piping and inter-process communication.
On macOS and Windows, the `--prevent-sleep` flag prevents the system from going idle while MCP Cron is running. It uses `caffeinate` on macOS and `SetThreadExecutionState` on Windows. Linux is not supported for sleep prevention.
It uses a local SQLite database. The default path is `~/.mcp-cron/results.db`, configurable with the `--db-path` flag or `MCP_CRON_STORE_DB_PATH` environment variable.
Use `--ai-provider` (openai or anthropic) and `--ai-model` flags, or their corresponding environment variables. API keys are set via `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or the generic `MCP_CRON_AI_API_KEY`. A custom base URL (`--ai-base-url`) supports OpenAI-compatible endpoints like LiteLLM.

A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code
AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
Comments