ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified productivity Safety 3/5

task-runner

Persistent task queue system. Users add tasks at any time via natural language; tasks are stored in a single persistent queue file and executed asynchronously via subagents. A heartbeat/cron dispatcher wakes periodically to check pending tasks, spawn workers, and report completions. The system never "finishes" — it always remains ready for the next task.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/chunhualiao/autonomous-task-runner
Or

What This Skill Does

The task-runner is a persistent, daemon-style task queue for OpenClaw. Unlike transient AI interactions, this skill creates a reliable infrastructure for handling multi-step workflows. It utilizes two operational modes: INTAKE and DISPATCHER. The INTAKE mode acts as the front-end, parsing natural language to extract, format, and serialize tasks into a persistent queue file. The DISPATCHER mode acts as the execution engine, polling the queue, spawning autonomous subagents to execute items, and managing the lifecycle of tasks—including retries and archival. Because it is designed to run asynchronously, the system is always ready, allowing you to offload complex or multi-part requests without waiting for immediate completion.

Installation

To add this capability to your agent, use the OpenClaw management utility. Run the following command in your terminal:

clawhub install openclaw/skills/skills/chunhualiao/autonomous-task-runner

After installation, define your environment preferences in your TOOLS.md file. You can configure the directory where tasks are stored (TASK_RUNNER_DIR), the level of concurrency (TASK_RUNNER_MAX_CONCURRENT), the number of retries before a task is considered blocked (TASK_RUNNER_MAX_RETRIES), and the cleanup threshold (TASK_RUNNER_ARCHIVE_DAYS).

Use Cases

  • Project Management: Break down a high-level project request into a series of actionable steps that the agent executes over time.
  • Asynchronous Automation: Batch operations that take a long time to complete (e.g., scraping, complex report generation, or multi-file refactoring).
  • Reliable Workflows: Use the retry logic to ensure that intermittent network failures or subagent errors do not permanently block your progress.
  • Delegation: Offload repetitive, multi-step maintenance tasks while you focus on higher-level problem solving.

Example Prompts

  1. "I need you to refactor the authentication module, update the unit tests for the login function, and then draft a summary of the changes in a new markdown file."
  2. "Add a task to check for broken links on the staging site, and then another task to archive the logs from last week."
  3. "Show me the current status of my task queue and retry any tasks that failed earlier today."

Tips & Limitations

For best results, phrase your task requests clearly with explicit action verbs. While the system supports complex compound asks, avoid adding too many unrelated tasks in a single message to ensure the subagents have clear context. Be aware that this skill performs file-write and code-execution operations; ensure your TASK_RUNNER_DIR is set to a secure, dedicated folder to prevent accidental file deletion or interference with other system processes. The system is intended to be "always on," so it may continue to consume compute resources via the heartbeat mechanism until explicitly disabled.

Metadata

Stars3562
Views13
Updated2026-03-29
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-chunhualiao-autonomous-task-runner": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#task-queue#automation#workflow#asynchronous#agent-delegation
Safety Score: 3/5

Flags: file-write, file-read, code-execution

Related Skills

git-repo-to-book

Write a full-length technical book using multi-agent AI orchestration. Spawns parallel research, writing, and review agents to produce 60K-100K+ word manuscripts. Also supports revising individual chapters of existing books. Based on the real workflow that produced an 88,000-word, 14-chapter book in under 18 hours.

chunhualiao 3562

claude-usage

Check Claude Max plan usage limits by launching Claude Code and running /usage. Use when the user asks about Claude plan usage, remaining quota, rate limits, or sends /claude_usage.

chunhualiao 3562

skill-engineer

Design, test, review, and maintain agent skills for OpenClaw systems using multi-agent iterative refinement. Orchestrates Designer, Reviewer, and Tester subagents for quality-gated skill development. Use when user asks to "design skill", "review skill", "test skill", "audit skills", "refactor skill", or mentions "agent kit quality".

chunhualiao 3562

rose-container-tools

Build and run ROSE compiler tools using ROSE installed in a Docker container. Use when developing source-to-source translators, call graph analyzers, AST processors, or any tool that links against librose.so. Triggers on "ROSE tool", "callgraph", "AST traversal", "source-to-source", "build with ROSE", "librose".

chunhualiao 3562

deepwiki

Query DeepWiki MCP to get AI-grounded answers about any public GitHub repository. Use when answering questions about a repo's source code, architecture, configuration, or internals. Triggers on "how does X work in <repo>", "deepwiki", "look up in codebase", "ask deepwiki", "check the source code".

chunhualiao 3562