A specialized maintenance skill that automates health monitoring, process cleanup, and performance optimization for autonomous agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-self-care
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 agent-self-care 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 agent-self-care?
Agent Self-Care is a dedicated utility designed to maintain the operational integrity of autonomous agents. By monitoring sub-agent status, process lifecycles, and context usage, this skill ensures that agents do not suffer from performance degradation due to stale sessions or resource leaks. It is a core component for anyone building resilient Openclaw Skills that need to run continuously without manual intervention.
The skill functions as a digital janitor, proactively cleaning up artifacts and rotating logs to prevent disk bloat. Its most powerful feature is the integration of BMAD-style retrospectives, which allow the agent to evaluate its own performance and document improvements in daily memory files, fostering a loop of continuous self-optimization.
agent-self-care Use Cases
- Running scheduled maintenance via cron to ensure long-term agent stability.
- Automatically killing sub-agents that have been unresponsive for more than 30 minutes.
- Managing context windows by triggering compaction when usage exceeds 80%.
- Rotating large log files and clearing temporary job artifacts to save space.
- Documenting daily performance gaps and improvements through automated retrospectives.
How agent-self-care Works
- The skill lists all active sub-agents and identifies stale sessions that have failed or exceeded the 30-minute progress threshold.
- It audits background processes, terminating any hanging tasks that haven't produced output in 5 minutes.
- Session health metrics are evaluated, specifically looking for context usage peaks or sessions older than 2 hours.
- An optimization shell script is executed to handle file-system level maintenance like log rotation.
- Every 10 runs, a retrospective is triggered to score the agent's effectiveness and update the memory directory.
agent-self-care Setup
Configure the agent to run the maintenance routine every 5 minutes by adding the following payload to your configuration for Openclaw Skills:
{
"name": "agent-self-care",
"schedule": {"kind": "every", "everyMs": 300000},
"payload": {"kind": "agentTurn", "message": "Run agent-self-care skill"},
"sessionTarget": "isolated",
"enabled": true
}
Ensure that the environment has the necessary permissions to execute the optimization script located at scripts/optimize.sh.
agent-self-care Data Schema & Taxonomy
The skill organizes its maintenance data and performance history into structured reports and markdown files.
| Component | Description |
|---|---|
| Self-Care Report | Console output detailing killed processes and health status. |
| Memory MD | Retrospective insights saved in memory/daily/YYYY-MM-DD.md. |
| Optimization Logs | System logs rotated and compressed once they exceed 50MB. |
| Health Metrics | Real-time tracking of context usage and token growth. |
agent-self-care Advanced Features
- Automated session status auditing with proactive compaction triggers.
- BMAD retrospective logic for continuous autonomous self-improvement.
- Integrated log rotation management for high-volume Openclaw Skills.
- Intelligent cleanup of completed cron job artifacts and temporary files.
- Performance scoring system to track agent efficiency over time.
SKILL.md
Loading
Related Openclaw Skills
Zoho Mail CLI
A robust command-line interface for Zoho Mail that outputs machine-readable JSON for seamless automation and agent workflows.
pm-sim
A high-fidelity simulation and paper trading environment for Polymarket prediction markets featuring 26 specialized MCP tools.
Duplicati Manager
A powerful interface to manage and automate Duplicati backup operations using secure Bearer tokens and REST API integration.
WeChat Moments 4-Grid Card Generator
Automatically transforms any topic into a complete WeChat Moments post featuring engaging copy and four professionally styled tutorial cards.
Cashu Emoji
A specialized tool for embedding and extracting Cashu tokens or hidden messages within standard emojis using Unicode variation selectors.
Clawtar
A demonstration skill designed for AI agents to practice detecting, settling, and retrying Cashu-gated HTTP 402 payment flows.
Featured*
