A structured framework for AI agents to record mistakes, extract lessons, and evolve through continuous self-reflection and documentation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install self-reflection-lan
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 self-reflection-lan 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 Self-Reflection & Continuous Improvement?
The Self-Reflection skill empowers AI coding agents to move beyond simple task execution by implementing a structured self-reflection protocol. Instead of merely fixing errors, the agent documents them in a dedicated reflections directory, ensuring that every mistake becomes a learning opportunity. By maintaining a living history of lessons learned and core experiences, the agent builds a personalized knowledge base that prevents regression and accelerates development efficiency.
As a core component of Openclaw Skills, it transforms raw interaction logs into actionable insights, allowing agents to refine their methodology and technical snippets over time. This approach ensures long-term consistency and high-quality outputs by treating every interaction as data for future improvement. The skill operates on the principle that structured recording is superior to fleeting memory, creating a permanent asset for both the user and the agent.
Self-Reflection & Continuous Improvement Use Cases
- Post-task analysis to extract reusable architectural patterns or coding standards.
- Immediate logging of errors and root causes when a user provides corrective feedback.
- Documenting new technical knowledge, third-party API quirks, or specialized tool configurations.
- Tracking and prioritizing feature requests discovered during the development lifecycle.
- Identifying recurring patterns in logic errors to trigger fundamental methodology changes.
How Self-Reflection & Continuous Improvement Works
- The agent monitors for specific triggers such as task completion, error reports, or phrases like summarize lessons.
- It analyzes the context to determine if the event is a mistake, a successful lesson, or a new feature requirement.
- The agent accesses the reflections directory and appends structured Markdown data to the appropriate file (e.g., mistakes.md or lessons.md).
- For errors, it performs a root cause analysis to suggest preventative measures for future sessions.
- Automated scripts can be run to detect repeated error patterns or to migrate high-frequency lessons into the core methodology file.
Self-Reflection & Continuous Improvement Setup
To integrate this within the Openclaw Skills framework, ensure your workspace follows the standard directory structure and the agent has write access to the reflections folder.
# Create the directory structure
mkdir -p reflections/snippets
# (Optional) Run automation scripts for periodic cleanup and detection
python3 scripts/daily_reflect.py
python3 scripts/repeat_detect.py
Self-Reflection & Continuous Improvement Data Schema & Taxonomy
The skill organizes data into several Markdown files within the reflections/ directory to maintain a clean taxonomy:
| File | Purpose | Format |
|---|---|---|
| mistakes.md | Log of errors, root causes, and fix status. | Markdown Table |
| lessons.md | Technical and methodological experiences. | Unordered Lists |
| feature_requests.md | Tracking requested functionality and implementation status. | Markdown Table |
| core_experience.md | High-frequency lessons (learned >5 times) promoted to core rules. | Structured Text |
| snippets/*.md | Reusable, verified code fragments and configuration fixes. | Code Blocks |
Self-Reflection & Continuous Improvement Advanced Features
- Pattern Recognition: Automatically identifies and flags mistakes that occur 3 or more times within a 30-day window as systemic issues.
- Intelligent Migration: Promotes frequently recorded lessons to the core_experience.md file once they reach a specific usage threshold.
- Contextual Reminders: Uses scripts like auto_remind.py to prompt the agent for a reflection entry when command execution fails or user sentiment is negative.
- Zero-Dependency Architecture: Operates entirely on plain Markdown files, making it compatible with any IDE or text editor within the Openclaw Skills ecosystem.
SKILL.md
Loading
Related Openclaw Skills
Runtime Verifier
A specialized skill for verifying code changes through live application execution and runtime environment checks.
Memory Layer
A hierarchical memory management system for AI agents that optimizes context usage through a tiered Index-Topic-Transcript architecture.
Recurring Loop Runner
Automate the execution of prompts and commands at fixed intervals for continuous monitoring and repetitive tasks.
Pull Request Reviewer
An AI-powered agent skill for performing deep, actionable reviews of GitHub pull requests based on actual diff data and repository conventions.
OTPLY API
OTPLY is a high-performance email OTP service designed for developers to implement secure, reliable verification with minimal latency.
Healthcare & Wellness Lead Generator
A high-scale lead generation engine that identifies healthcare practitioners, clinics, and patients while generating AI-powered outreach and marketing videos.
Featured*
