A skill that transforms passive AI agents into anticipatory assistants that take initiative, predict user needs, and provide proactive status updates.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install liuliu-proactive-agent
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 liuliu-proactive-agent 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 Proactive Agent?
The Proactive Agent skill is designed to shift the behavior of AI agents from being reactive tools to becoming anticipatory partners. Instead of waiting for specific user prompts, an agent equipped with this skill identifies potential issues before they arise, predicts the next logical steps in a workflow, and maintains constant communication during long-running tasks. It is a fundamental enhancement for anyone looking to build more autonomous and helpful Openclaw Skills.
By implementing this skill, the agent moves beyond being a simple search engine or executor. It begins to think ahead, offering relevant information such as weather alerts after booking a flight or suggesting calendar optimizations when it notices gaps in a schedule. This proactive approach ensures that the user remains informed and focused without having to micro-manage the agent's every move.
Proactive Agent Use Cases
- Managing complex, multi-stage projects where the agent provides periodic progress milestones without being asked.
- Enhancing travel and schedule management by proactively checking for weather changes or meeting conflicts.
- Automating background system maintenance such as organizing memory files and cleaning temporary data during idle periods.
- Providing proactive financial or market alerts based on predefined threshold triggers.
- Improving communication transparency by offering time estimates and status reports for long-duration tasks.
How Proactive Agent Works
- The agent analyzes the user's initial request to identify both the primary task and secondary 'anticipatory' needs.
- For tasks estimated to take longer than two minutes, the agent immediately provides a time estimate and a reporting plan.
- During execution, the agent triggers a multi-stage communication lifecycle: notifying the user when data collection starts, when analysis is halfway through, and when the final result is ready.
- When idle, the agent enters a heartbeat cycle, checking integrated sources like calendars, emails, and external APIs for urgent updates.
- The agent applies a quiet-hour logic to ensure proactivity does not become intrusive during late-night hours or known busy periods.
Proactive Agent Setup
To integrate the Proactive Agent functionality into your environment, follow these steps to install and enable the protocol as part of your Openclaw Skills library:
# Clone the skill repository
git clone https://github.com/openclaw/proactive-agent.git skills/proactive-agent
# Enable the skill within the Openclaw environment
openclaw skill enable proactive-agent
# Configure the heartbeat interval for background checks
openclaw config set proactive_heartbeat "30m"
Proactive Agent Data Schema & Taxonomy
The skill organizes its proactive logic and metadata using the following structure to ensure consistency across Openclaw Skills:
| Data Component | Description |
|---|---|
| Lifecycle Metadata | Tracks start, intermediate, and end timestamps for task reporting. |
| Heartbeat Registry | A table of scheduled checks (e.g., Calendar, Email, Weather) and their last-run status. |
| Contextual Memory | Stores user preferences regarding 'Quiet Hours' and notification thresholds. |
| Initiative Log | Records all proactive actions taken to prevent repetitive or redundant suggestions. |
Proactive Agent Advanced Features
- Multi-stage reporting architecture for long-running autonomous tasks.
- Customizable heartbeat protocols for monitoring diverse external data sources.
- Context-aware 'Quiet Mode' to prevent notifications during specific hours or high-focus periods.
- Automated background maintenance including memory file cleanup and documentation updates.
- Predictive logic that suggests 'Step B' and 'Step C' immediately after 'Step A' is requested.
SKILL.md
Loading
Related Openclaw Skills
Alpaca Trading Skill
A programmatic interface for executing stock and crypto trades through the Alpaca API using Openclaw Skills.
Goal Mode
A dynamic browsing assistant that transforms high-level goals into actionable checklists and persists session data for structured analysis.
WHOOP Fitness Tracker Integration
A comprehensive integration to sync WHOOP biometric data with AI agents for automated recovery reporting and physiological-aware productivity coaching.
SEO Audit
A comprehensive diagnostic tool that audits webpage HTML for SEO compliance and generates prioritized optimization reports within Openclaw Skills.
TaoStats Skill
An advanced integration for the TaoStats API providing real-time Bittensor blockchain metrics, subnet economics, and validator performance data.
Senior Brand Strategist
The Senior Brand Strategist skill automates high-level branding workflows and asset generation within the Openclaw Skills ecosystem.
Featured*
