Daily Evolution for Openclaw

An automated self-iteration engine that enables AI agents to learn from daily interactions, optimize toolchains, and refine their own core persona files.

wilsonliu95
v1.0.0
Feb 7, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install daily-evolution

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install daily-evolution 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 Daily Evolution?

The Daily Evolution engine serves as a continuous self-improvement framework for AI agents, allowing them to transition from static tools to evolving collaborative partners. By analyzing daily interaction logs, the engine extracts high-efficiency patterns and identifies potential failures to ensure the agent's behavior aligns more closely with user preferences over time. This skill is essential for developers looking to build resilient, self-optimizing agents within the ecosystem of Openclaw Skills.

Operating on a four-phase lifecycle—sorting, inventory, solidification, and reporting—this engine ensures that every interaction contributes to a smarter, more capable agent. It systematically updates core persona documents and tool configurations, ensuring that best practices are not just discovered but permanently integrated into the agent's operational logic.

Daily Evolution Use Cases

  • Automating the daily performance review of AI agents to identify collaboration bottlenecks.
  • Continuously updating agent personas (SOUL.md) based on real-world user feedback and habits.
  • Maintaining a healthy toolchain by automatically auditing MCP services and API key validity.
  • Hardening temporary insights into long-term memory to prevent context loss and improve task execution quality.

How Daily Evolution Works

  1. Conversation Sorting: The agent reviews the previous 24 hours of dialogue to extract patterns, user preferences, and high-impact collaboration techniques.
  2. Tool Inventory: A comprehensive audit of all installed tools and MCP services is performed to verify connectivity, API status, and usage frequency.
  3. Best Practice Solidification: Validated insights are written into core configuration files like SOUL.md, TOOLS.md, and MEMORY.md using clear version diffs.
  4. Execution Reporting: The engine generates a structured markdown report in the workspace/reports directory summarizing the day's evolution and setting objectives for the next cycle.

Daily Evolution Setup

To enable the self-evolution cycle, configure a cron job to trigger the engine automatically.

# Manual trigger command
openclaw run daily-evolution

Add the following configuration to your environment settings to automate the process:

{
  "id": "daily-evolution",
  "schedule": "0 0 * * *",
  "timezone": "Asia/Shanghai",
  "prompt": "Execute daily-evolution skill. Follow the four-phase cycle: sort, inventory, solidify, report."
}

Daily Evolution Data Schema & Taxonomy

The skill manages and generates data according to the following structure:

Data Component Format Description
Evolution Reports reports/evolution-{date}.md Daily summaries of insights, tool health, and persona updates.
Persona (SOUL.md) Markdown Stores evolving collaboration styles, values, and user habits.
Heartbeat (HEARTBEAT.md) Markdown Strategy for recurring tasks and health check frequencies.
Tool Memory (TOOLS.md) Markdown Records best practices, usage tips, and known issues for integrated tools.
Long-term Memory MEMORY.md Distilled archive of critical historical interactions and decisions.

Daily Evolution Advanced Features

  • Automated cron integration for zero-touch daily maintenance and self-upgrading.
  • Multi-file persona management allowing granular updates to values, memory, and tool logic.
  • Sophisticated insight extraction that filters one-off preferences from repeatable best practices.
  • Built-in security guardrails that prevent the agent from modifying project code or sensitive user data during evolution.
  • Integrated health monitoring for Openclaw Skills and MCP extensions to ensure operational continuity.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*