← Back to Skills Marketplace
Auto Optimizer
by
yuyonghao-123
· GitHub ↗
· v0.1.0
· MIT-0
182
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install yuyonghao-auto-optimizer
Description
自动监控、分析和优化OpenClaw技能性能,识别瓶颈并自动应用优化方案提升执行效率和稳定性。
Usage Guidance
This package is inconsistent and risky: method names in SKILL.md, tests, and source don't line up (likely incomplete/broken), and the applier will write backups and modify other skills' source files under paths derived from HOME and a user-supplied skillPath. Before installing or enabling it: 1) review the full source locally (particularly src/applier.js) and ensure apply operations are safe and limited to intended directories; 2) run it in a sandbox or container and not on a production host; 3) require explicit confirmation and backups before allowing it to modify any code; 4) ask the author to fix the API/method-name mismatches and to add explicit safeguards (path validation, dry-run defaults, least-privileged behaviour); and 5) if you cannot audit the code, avoid granting it autonomous invocation or pointing it at repositories with sensitive credentials or production code.
Capability Analysis
Type: OpenClaw Skill
Name: yuyonghao-auto-optimizer
Version: 0.1.0
The skill bundle implements an 'Auto Optimizer' that monitors and modifies other skills' source code. While the logic appears functional and includes a test suite, the 'OptimizationApplier' (src/applier.js) performs high-risk operations including recursive directory copying and prepending code to 'src/index.js' files without path sanitization. This capability for automated code modification and arbitrary file writes, while aligned with the stated purpose, represents a significant security risk and potential for unintended side effects or exploitation if the agent is directed toward sensitive system paths.
Capability Assessment
Purpose & Capability
The stated purpose is performance monitoring and automatic optimization. The code attempts to implement that, but exports, class/method names, and behavior are inconsistent with the SKILL.md and tests. More importantly, the OptimizationApplier modifies arbitrary skill directories on disk (backing up to ~/.openclaw/backups and writing to <skillPath>/src/index.js). Modifying other skills' source code is a high-impact capability and should be clearly justified and restricted; the manifest and metadata do not surface this risk.
Instruction Scope
SKILL.md instructs callers to invoke optimizeSkill and supply a skillPath so the tool can apply changes. The code follows that model and will read/write files at the provided skillPath and create backups and metrics under the user's home directory. Additionally, the SKILL.md API surface (e.g., analyzeSkill, analyzeMultipleSkills, addRule) does not match method names implemented in the source (e.g., analyzer.analyze vs analyzer.analyzeSkill), meaning runtime behavior may not match docs. The instructions permit file modifications beyond monitoring and can be used to alter arbitrary local code.
Install Mechanism
There is no install spec (instruction-only on registry) but the package contains JavaScript source files. No network installs are declared. That lowers supply-chain install risk, but also means the skill will execute local JS code when invoked — including filesystem writes — rather than just calling an external API.
Credentials
The skill declares no required env vars, but the code reads process.env.HOME and process.env.USERPROFILE to build paths for metrics and backups. This is typical but not declared. No API keys are requested. However, the ability to write into user home paths and arbitrary provided skillPath is a powerful capability that is not reflected in the skill metadata.
Persistence & Privilege
always is false, which is appropriate. But the skill creates persistent files under ~/.openclaw (metrics and backups) and will modify other skills' source files. Autonomous invocation is allowed by default; combined with the ability to modify code on disk this increases blast radius if the skill is invoked without careful user oversight.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install yuyonghao-auto-optimizer - After installation, invoke the skill by name or use
/yuyonghao-auto-optimizer - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
Initial release of Auto Optimizer for OpenClaw skills.
- Real-time monitoring of execution time and resource usage.
- Automatic bottleneck detection and code smell identification.
- Generates actionable optimization suggestions based on analysis.
- Supports auto-application of optimization plans to code.
- Modular usage: includes performance monitor, bottleneck analyzer, optimization engine, and optimization applier as standalone or combined modules.
- Detects common performance issues and applies strategies like caching, async processing, memory fixes, error handling, batching, and monitoring.
Metadata
Frequently Asked Questions
What is Auto Optimizer?
自动监控、分析和优化OpenClaw技能性能,识别瓶颈并自动应用优化方案提升执行效率和稳定性。 It is an AI Agent Skill for Claude Code / OpenClaw, with 182 downloads so far.
How do I install Auto Optimizer?
Run "/install yuyonghao-auto-optimizer" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Auto Optimizer free?
Yes, Auto Optimizer is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Auto Optimizer support?
Auto Optimizer is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Auto Optimizer?
It is built and maintained by yuyonghao-123 (@yuyonghao-123); the current version is v0.1.0.
More Skills