A sophisticated task queue system that filters autonomous agent work by type, ensuring tokens are spent on high-value activities while routine maintenance is handled by scripts.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install autonomy-type-based
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install autonomy-type-based using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Type-Based Autonomy Task Queue?
The Type-Based Autonomy skill transforms a reactive AI agent into a focused autonomous worker by implementing a structured task classification system. Instead of processing every item in a list, the agent intelligently filters tasks/QUEUE.md to identify tasks tagged with specific high-value types like research, writing, or analysis. This approach is a pillar for developers building robust Openclaw Skills that require a clear separation between intelligent reasoning and routine system operations.
By utilizing this skill, you can ensure that your agent ignores low-value maintenance tasks—such as backups or log cleanups—which are better suited for scheduled cron jobs. This maintains a clean operational environment while maximizing the ROI of your LLM token usage. It provides a scalable framework for task management that grows with your agent's capabilities.
Type-Based Autonomy Task Queue Use Cases
- Scaling autonomous workflows where research and content generation must be prioritized over system upkeep.
- Implementing token-efficient agent behaviors by restricting autonomy to specific domains.
- Managing complex multi-task environments where human-assigned priorities dictate the agent's focus.
- Coordinating work between autonomous agents and automated system scripts (cron).
How Type-Based Autonomy Task Queue Works
- The agent performs a heartbeat check and reads the central
tasks/QUEUE.mdfile. - It filters the 'Ready' section for specific labels such as
@type:research,@type:writing, or@type:analysis. - The system ignores any tasks tagged with
@type:maintenance,@type:backup, or@type:security, leaving those for external automation. - The agent selects the highest priority task based on the
@prioritylabel (Urgent > High > Medium > Low). - Work is executed, and the agent updates the queue by moving the task to 'In Progress' and eventually 'Done Today'.
- Insights are logged to the
.learnings/directory and progress is cross-referenced withGOALS.md.
Type-Based Autonomy Task Queue Setup
To integrate this logic into your Openclaw Skills environment, ensure your directory structure is prepared:
# Create necessary directories
mkdir -p tasks/archive
mkdir -p .learnings
# Initialize your task queue
touch tasks/QUEUE.md
Ensure your agent's system prompt is updated to recognize the @type and @priority taxonomy defined in the skill documentation.
Type-Based Autonomy Task Queue Data Schema & Taxonomy
The skill relies on a standardized Markdown schema within the tasks/ directory:
| Element | Format | Purpose |
|---|---|---|
| Task File | tasks/QUEUE.md |
The primary manifest for all agent activity. |
| Task Types | @type:[category] |
Categorizes tasks (research, writing, analysis, maintenance). |
| Priority | @priority:[level] |
Determines execution order (urgent, high, medium, low). |
| Learnings | .learnings/LRN-*.md |
Persistent storage for findings and insights. |
| Output | tasks/outputs/*.md |
Location for generated reports and content. |
Type-Based Autonomy Task Queue Advanced Features
- Strategic Token Budgeting: Recommends session-based token limits to prevent runaway costs during autonomous loops.
- GOALS.md Alignment: Forces the agent to validate every task against long-term objectives before execution.
- Cron Parallelism: Enables a 'hybrid' architecture where the agent and system scripts work in the same queue without conflicts.
- Dynamic Idea Discovery: The agent can autonomously identify follow-up tasks and add them to an 'Ideas' section for future review.
SKILL.md
Loading
Related Openclaw Skills
Notion Tasks Blocks
A specialized skill for managing top-level to-do blocks in Notion pages without the overhead of databases.
Notion MVP Task Manager
A high-performance integration for managing Notion tasks, agendas, and database entries through simple CLI commands and AI workflows.
gogcli
A comprehensive command-line interface for managing Google Workspace services including Gmail, Drive, and Calendar directly from your terminal.
Supabase
A comprehensive toolkit for Supabase database management and vector similarity search within Openclaw Skills.
Windowed Autonomy
A time-windowed autonomous task queue that separates active daytime work from overnight maintenance cycles for predictable agent behavior.
Weather Checker
A professional Python-based command-line utility providing global weather forecasts, precipitation data, and temperature tracking with emoji-rich formatting.
Featured*
