← Back to Skills Marketplace
464169
Downloads
3839
Stars
7019
Active Installs
34
Versions
Install in OpenClaw
/install self-improving-agent
Description
A skill where the agent logs it's own findings for self-improvement
Usage Guidance
Before installing, decide whether you want only manual logging or also automatic reminders. Review hook scripts before enabling them, avoid global hooks unless you want reminders in every session, and do not let the skill save secrets, raw transcripts, or full command output into persistent learning files.
Capability Assessment
Purpose & Capability
The core purpose is coherent: it logs errors, corrections, feature requests, and reusable learnings to local markdown files. Optional hook reminders, OpenClaw integration, and skill scaffolding expand the workflow but remain related to learning capture.
Instruction Scope
The documentation recommends broad prompt-time reminders and optional Bash-output error detection. These are disclosed and opt-in, but users should understand they can influence many sessions if configured globally.
Install Mechanism
Installation is conventional for an OpenClaw skill, with manual clone/copy options and a separately enabled hook. No automatic or hidden install-time persistence was found in the artifacts.
Credentials
The skill writes persistent local learning files and may update agent memory files when the user follows the workflow. The artifacts repeatedly advise avoiding secrets, raw transcripts, and full command output.
Persistence & Privilege
Persistence is limited to local markdown learning files, optional workspace guidance files, and optional hooks. The scripts inspected do not perform network access, credential collection, or destructive actions; the scaffold helper only creates a validated relative skill directory.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install self-improving-agent - After installation, invoke the skill by name or use
/self-improving-agent - Provide required inputs per the skill's parameter spec and get structured output
Changelog
- Improved skill documentation for self-improvement logging, including detailed setup instructions for OpenClaw and generic agents.
- Clarified initialization requirements: ensure `.learnings/` directory and key markdown files exist before use.
- Added comprehensive quick-reference table mapping common situations (errors, corrections, requests) to specific log files.
- Expanded promotion guidance for learnings, with clear targets for AGENTS.md, SOUL.md, TOOLS.md, and project memory.
- New, structured markdown templates for logging learnings and errors, focusing on concise, non-sensitive summaries.
- Guidance for integrating with OpenClaw sessions and optional auto-reminder hook workflow.
Metadata
Frequently Asked Questions
What is self-improving agent?
A skill where the agent logs it's own findings for self-improvement. It is an AI Agent Skill for Claude Code / OpenClaw, with 464169 downloads so far.
How do I install self-improving agent?
Run "/install self-improving-agent" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is self-improving agent free?
Yes, self-improving agent is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does self-improving agent support?
self-improving agent is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created self-improving agent?
It is built and maintained by pskoett (@pskoett); the current version is v3.0.24.
More Skills