A dedicated agent skill designed to automatically modify, update, and optimize your local agent workflows based on real-world execution feedback.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-optimizer-guipi
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 skill-optimizer-guipi 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 Skill Optimizer?
The Skill Optimizer is a robust, meta-utility built to continuously maintain and refine your personal library of Openclaw Skills. Over time, static AI agent prompt guidelines and step-by-step procedures can drift or run into unexpected real-world edge cases. Instead of manually editing text files when workflows break, this tool serves as a self-healing layer that steps in whenever you discover loopholes, missing workflows, or logical errors.
By leveraging this tool within your automation loop, your developer agent can dynamically rewrite target SKILL.md guidelines, update semantic version tags, and catalog execution failures. This ensures that every mistake becomes a permanent, codified lesson, making your collection of Openclaw Skills progressively smarter, more resilient, and perfectly adapted to your specific computing environment.
Skill Optimizer Use Cases
- Operational Pitfall Codification: Instantly register operational errors and build preventative checks into your workflows so your agent never repeats the same mistake.
- Dynamic Workflow Appending: Seamlessly add new parameters, files, or checklist steps to an existing skill when operational requirements change.
- Schedules & Rule Iteration: Quickly adjust execution intervals, rule conditions, or validation thresholds inside your agent documents using simple, conversational commands.
- Continuous Prompt Refinement: Automatically optimize prompt structures, reducing token usage and refining instructions within target Openclaw Skills for enhanced processing accuracy.
How Skill Optimizer Works
- Identify Target Skill: The optimizer parses user instructions to locate the exact skill directory and its SKILL.md file inside your workspace.
- Diagnose Issue Type: It classifies the requested change into clear categories such as loopholes, rule updates, missing workflow steps, or pitfall entries.
- Inspect Active Settings: The agent reads and analyzes the existing markdown structure of the target skill.
- Apply Minimal Diff: Adhering strictly to the principle of minimal change, it surgically injects updates to preserve original configurations while applying the modification.
- Version Incrementation: The skill's version is automatically bumped, and a concise summary is written to the iteration history log.
- Pitfall Cataloging: If the modification addresses a runtime mistake, the optimization details are cross-referenced and logged in the local pitfalls registry.
Skill Optimizer Setup
To deploy and configure the Skill Optimizer within your local directory of Openclaw Skills, make sure your workspace follows the expected structural layout.
Directory Configuration
Your local directory should be organized as follows:
~/.workbuddy/skills/
└── guipi-skill-optimizer/
├── SKILL.md
└── references/
├── 修改规范.md (Modification Rules)
├── 踩坑总库.md (Consolidated Pitfalls Database)
└── 优化经验.md (Proven Optimization Practices)
CLI Quick Operations
Use these standard commands in your terminal to inspect and audit your skills:
# Inspect the contents of a target skill
cat ~/.workbuddy/skills/{skill_name}/SKILL.md
# Audit historical pitfall logs for a specific skill
cat ~/.workbuddy/skills/{skill_name}/references/pitfalls.md
# List all registered skills in your environment
ls ~/.workbuddy/skills/ | sort
Skill Optimizer Data Schema & Taxonomy
The Skill Optimizer enforces a clear markdown-based taxonomy across all Openclaw Skills to keep documentation tidy and highly searchable. Below is the file structure schema utilized by this optimizer:
| File Name / Path | Format / Schema | Primary Role |
|---|---|---|
SKILL.md |
YAML Frontmatter + Markdown | Houses metadata (slug, version, triggers), execution workflows, and change history logs. |
references/修改规范.md |
Markdown Markdown Guidelines | Enforces the strict "minimal change" standard and formatting patterns for updating skills. |
references/踩坑总库.md |
Tabular Markdown | Holds a global, cross-skill directory of edge cases and logic failures. |
references/优化经验.md |
Markdown Cheat Sheet | Lists proven prompt engineering patterns that maximize agent productivity. |
Skill Optimizer Advanced Features
- Principle of Minimal Change: Surgically modifies files to preserve the target prompt's structure, preventing bulk overwrites and losing custom nuances.
- Global Pitfalls Database Integration: Automatically maintains cross-skill registries (
踩坑总库.md), enabling lessons learned in one domain to safeguard other Openclaw Skills. - Automated Version Tracking: Calculates and executes incremental version upgrades, maintaining structured and accurate changelogs with every execution loop.
- Adaptive Rule Parsing: Intelligently maps informal user complaints or task reviews directly into technical constraints inside the core agent guidelines.
SKILL.md
Loading
Related Openclaw Skills
Debug TypeScript Error
A local-first, standalone engineering skill designed to translate complex TypeScript errors into plain English and implement surgical, safe type fixes without relying on lazy type suppression.
Generate Storybook Story
An AI-powered local development skill that automatically generates clean, minimal, and highly realistic Storybook stories for frontend components.
API Change Review
An automated development skill that reviews API diffs for compatibility, security, and schema evolution, classifying changes as additive, behavior-changing, or breaking.
Performance Regression Triage
A structured workflow for AI agents to systematically isolate and triage system performance issues using concrete telemetry data.
Database Migration Risk Review
A local-first AI coding agent skill designed to classify database migrations, expose runtime production risks, and recommend zero-downtime deployment sequences.
Mac Disk Audit Advisor
A secure, developer-centric macOS storage auditing and optimization tool that scans disk usage, detects massive files, and generates actionable, non-destructive cleaning commands.
Featured*
