Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Agents Cron

    1

    Automate recurring agent tasks using cron schedules and a local CLI-driven task store.

    Free

    1 useSecurity scanned
    agents-cron

    agents-cron

    Example session with this skill installed

    Schedule a task to run every day at 9am that checks for any TODO comments in the src directory and summarizes them in a file named TODO_REPORT.md.

    • Read your context and instructions
    • Compiled the agents-cron

    I have scheduled a new recurring task for your project. This task will run daily using the default agent adapter.

    Scheduled Task Created

    FieldValue
    Task IDt-8f2k9l3
    PromptCheck for TODO comments in the src directory and summarize them in TODO_REPORT.md.
    Schedule0 9 * * * (Daily at 09:00)
    Agentpi
    StatusEnabled
    Next Run2023-11-25 09:00:00
    Store Path.agents/cron/tasks.json

    Next steps

    • Ensure you have agents-cron run-due configured in your system crontab to trigger the execution.
    • You can view the status of this task at any time by running agents-cron dashboard.
    • To test the prompt immediately without waiting for the schedule, run agents-cron run t-8f2k9l3.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Standard AI agents are reactive, waiting for a prompt before they act. This skill enables scheduled and recurring agent execution within your local development environment, letting you automate code audits, documentation updates, or status reports without manual intervention.

    What it does

    • Schedule prompt execution by storing prompts with specific time rules for later or recurring runs.
    • Manage task lifecycle through commands to add, enable, disable, or remove tasks in a project-scoped store.
    • Inspect run history via a local dashboard that tracks summaries, errors, and session records for every execution.
    • Multi-agent support allows tasks to run through Pi, Claude Code, or Codex adapters.
    • Granular time control supports ISO timestamps, duration intervals, and standard cron expressions.

    How it works

    1. Define the task using commands like /loop, /remind, or /schedule followed by your prompt and timing.
    2. Assign an agent to handle the task, such as pinning a headless Claude Code session for code-intensive work.
    3. External trigger occurs when your system's scheduler calls the CLI to execute due tasks.
    4. Monitor via dashboard by launching the local web view to debug failures or review successful completions.

    Frameworks & tools

    Built for Node.js (v22.19+) environments. Integrates with the agents-cron CLI and supports agent adapters for Pi, Claude Code, and Codex.

    Why this beats prompting it yourself

    Manually running the same maintenance prompts every morning is a waste of developer time. This skill provides a structured persistence layer and CLI for prompts, ensuring background tasks run reliably and providing a centralized dashboard to audit the results across multiple agent types.

    Use cases

    • Schedule a daily code quality audit every morning at 8 AM.
    • Remind an agent to check for outdated dependencies every Monday.
    • Run a recurring documentation sync between source code and README files.
    • Automate a project status summary for delivery every Friday afternoon.

    Known limitations

    Execution requires an external scheduler (like system cron) to invoke the CLI. Headless Claude Code tasks require an explicit tool allowlist to bypass interactive permission prompts.

    Requires Node.js 22.19+. Does not run tasks in the background autonomously; needs an external trigger like system cron to invoke the CLI. Interactive confirmation required for task deletion.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 29 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Write Files
    Read Files
    Network Access
    Browser
    Listed29 days ago

    What's inside

    Frequently Asked Questions

    While agents are typically reactive, this skill allows you to transform them into proactive workers by storing prompts in a local task store and executing them based on cron schedules or time intervals.

    The skill includes adapters for Pi, Claude Code, and Codex, allowing you to route specific scheduled tasks to the agent best suited for the job.

    You will receive the scheduler logic, the local CLI-driven task store, and the dashboard for monitoring run history; however, you must provide the external trigger (like system cron) to invoke the CLI.

    Yes, it requires Node.js (v22.19+) and is designed for local development environments where you can manage a project-scoped task store.

    The skill includes a built-in local dashboard that allows you to inspect run history, view session records, and debug errors from previous executions.

    For agents like Claude Code, you will need to configure an explicit tool allowlist within your environment to ensure tasks run smoothly without waiting for manual interactive permissions.

    New in Workflow Automation