← Back to Skills Marketplace
200423
Downloads
1218
Stars
2231
Active Installs
22
Versions
Install in OpenClaw
/install self-improving
Description
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
Usage Guidance
Install only if you want the agent to keep cross-session local memory about your corrections, preferences, and workflow patterns. Review ~/self-improving/ periodically, avoid storing secrets or sensitive personal data, use strict/passive mode if you want more confirmation, and review the separate Proactivity skill before approving its optional installation.
Capability Analysis
Type: OpenClaw Skill
Name: self-improving
Version: 1.2.16
The skill implements a structured, tiered memory system designed to help the agent learn from user corrections and self-reflection. It manages local files in `~/self-improving/` and includes a robust 'Security Boundaries' document (`boundaries.md`) that explicitly forbids storing credentials, financial data, or PII. While the setup process involves modifying core workspace configuration files like `SOUL.md` and `AGENTS.md` to establish persistence and steering, these actions are transparently documented and strictly aligned with the stated purpose of improving agent performance over time.
Capability Assessment
Purpose & Capability
The stated purpose is persistent self-improvement through corrections, reflections, and preferences, which matches the local memory files, learning rules, and heartbeat maintenance artifacts.
Instruction Scope
The skill has broad learning triggers and automatic logging, but it also limits learning to corrections/repeated patterns, forbids sensitive categories, requires confirmation for permanent rules, and exposes audit/forget/export commands.
Install Mechanism
Setup creates files under ~/self-improving/ and may edit AGENTS.md, SOUL.md, and HEARTBEAT.md; those paths are disclosed, and the optional Proactivity skill install is explicitly gated on user agreement.
Credentials
No credentials, elevated privileges, required binaries, or default network calls are requested; file access is scoped to the self-improving directory plus disclosed workspace steering files during setup.
Persistence & Privilege
Persistent local memory and heartbeat state are central to the skill and clearly described; users should review the stored contents because they can influence later agent behavior.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install self-improving - After installation, invoke the skill by name or use
/self-improving - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.16
Clarifies the setup flow for proactive follow-through and safer installation behavior.
v1.2.15
Improves the setup flow for stronger follow-through and execution quality.
v1.2.14
Updated the title to better highlight proactive self-improvement.
v1.2.13
Aligns heartbeat installation with the standard workspace setup so recurring maintenance is added alongside the existing self-improving routing.
v1.2.12
Refined the description to make the activation trigger more explicit.
v1.2.11
Refined the title and description to clarify proactive self-improvement triggers.
v1.2.10
Sharper setup now lists relevant memory before non-trivial work, with a title that highlights proactive self-reflection.
v1.2.9
Refined AGENTS.md memory-routing wording for clarity while preserving behavior.
v1.2.8
Refined AGENTS.md patch guidance to route corrections and lessons to self-improving by default, with AGENTS/TOOLS only for global rules.
v1.2.7
Added non-destructive AGENTS.md memory routing guidance to reinforce factual vs self-improving storage.
v1.2.6
Added a SOUL.md setup snippet to reinforce pre-task review and post-response learning compounding.
v1.2.5
Restored self-learning and self-organizing memory description while keeping setup behavior unchanged.
v1.2.4
Setup now triggers only when memory storage is missing, keeping activation flow cleaner.
v1.2.3
Turned self-improving into a true always-on loop: pre-task activation, post-response reflection, and tighter continuous learning behavior.
v1.2.2
Turned self-improving into a true always-on loop: pre-task activation, post-response reflection, and tighter continuous learning behavior.
v1.2.1
Clarified the core promise to highlight auto-learning from corrections and self-organizing memory for continuous improvement.
v1.1.3
Fixed skill title display.
v1.1.2
- Added EXTRA_FILES.txt to the project.
- No changes to functionality or user-visible features.
v1.1.1
Added self-reflection capability — agent now evaluates its own work, not just user corrections.
v1.2.0
Added self-reflection loop, experience-based learning, and visual workflow diagram.
Metadata
Frequently Asked Questions
What is Self-Improving + Proactive Agent?
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when... It is an AI Agent Skill for Claude Code / OpenClaw, with 200423 downloads so far.
How do I install Self-Improving + Proactive Agent?
Run "/install self-improving" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Self-Improving + Proactive Agent free?
Yes, Self-Improving + Proactive Agent is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Self-Improving + Proactive Agent support?
Self-Improving + Proactive Agent is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).
Who created Self-Improving + Proactive Agent?
It is built and maintained by Iván (@ivangdavila); the current version is v1.2.16.
More Skills