← Back to Skills Marketplace
171
Downloads
0
Stars
3
Active Installs
1
Versions
Install in OpenClaw
/install optimize
Description
Optimize OpenClaw performance and prevent lag. Use when: (1) OpenClaw feels slow or laggy, (2) High memory usage, (3) Slow response times, (4) Gateway crashe...
Usage Guidance
This package appears to do what it claims (optimize OpenClaw) but includes actions that can delete your OpenClaw history and affect system caches and running Python state. Before installing or running: 1) Back up ~/.openclaw/workspace and your vector DB; 2) Inspect the bundled scripts (openclaw_optimize_pro.py, openclaw_optimize.py, openclaw-optimize.sh) yourself—note the code deletes files and writes to /proc; 3) Avoid running as root — if you must run privileged actions (drop_caches), do so deliberately and on a test system; 4) Prefer running first in an isolated/staging environment or container; 5) If you only want diagnostics, run read-only functions (get_status, analyze_load_time) rather than commands that call clean_old, aggressive_gc (which clears sys.modules), or clear_cache; 6) If you trust the author, still consider adding a dry-run option or modifying the scripts to prompt before destructive actions.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-optimize-release-v2-0-0-tar
Version: 2.0.0
The skill bundle provides performance optimization tools that perform high-risk operations, including deleting files in the OpenClaw workspace (HistoryOptimizer), modifying system-level kernel parameters like /proc/sys/vm/drop_caches (MemoryOptimizerPro), and automatically installing Python dependencies via pip (openclaw-optimize.sh). While these actions are aligned with the stated purpose of 'optimization' and 'cleaning,' they involve significant system and data risks. Additionally, the use of sys.modules.clear() in openclaw_optimize_pro.py is an unconventional and potentially disruptive approach to memory management. No clear evidence of malicious intent or data exfiltration was found, but the broad permissions and aggressive system modifications warrant a suspicious classification.
Capability Assessment
Purpose & Capability
Name/description match the actual capabilities: memory optimization, skill-load analysis, history cleanup, monitoring and config tweaks. The code operates on ~/.openclaw workspace and enumerates skills, which is proportionate to optimizing OpenClaw.
Instruction Scope
Runtime instructions and bundled scripts perform file deletions (clean_old removes files from the memory directory and archive operations), manipulate a local vector DB, call into /proc/sys/vm/drop_caches (system-wide cache clearing), call sync, and the Pro code clears sys.modules. These actions can delete user data, require elevated privileges, and can destabilize running Python environments. The SKILL.md and shell scripts also run pip installs and suggest git/wget installs — expected, but destructive actions are present.
Install Mechanism
This is instruction-only with bundled code files; no opaque remote installers are executed by the skill itself. README suggests downloads from GitHub releases (a common host). No evidence of URL shorteners or untrusted hosts in the provided install instructions.
Credentials
The skill requests no environment variables or external credentials. It does, however, access and modify local OpenClaw data paths (~/.openclaw/workspace) and system cache interfaces — capabilities that are relevant to optimization but are powerful and should be granted deliberately.
Persistence & Privilege
The skill is not always: true and does not request elevated platform privileges directly, but its operations (writing to /proc/sys/vm/drop_caches, deleting files in workspace, clearing Python modules) require elevated permissions or can have system-wide effects. It modifies local OpenClaw workspace files (within scope) but also performs system-level cache clearing which is out-of-skill-scope for unprivileged installs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install optimize - After installation, invoke the skill by name or use
/optimize - Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.0
openclaw-optimize v2.0.0 introduces a comprehensive performance tuning toolkit for OpenClaw:
- Adds commands and Python API for automatic optimization, including memory management, skill analysis, history cleaning, and performance monitoring.
- Provides detailed usage scenarios, troubleshooting, and fix recipes for deployment lag, slowdowns, and gateway issues.
- Includes ready-to-use configuration templates for high performance and memory optimization.
- Features integrated tools for diagnostics, benchmarking, and automated scheduled optimization tasks.
- Chinese-language guides and best practices for rapid adoption.
Metadata
Frequently Asked Questions
What is openclaw-optimize?
Optimize OpenClaw performance and prevent lag. Use when: (1) OpenClaw feels slow or laggy, (2) High memory usage, (3) Slow response times, (4) Gateway crashe... It is an AI Agent Skill for Claude Code / OpenClaw, with 171 downloads so far.
How do I install openclaw-optimize?
Run "/install optimize" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is openclaw-optimize free?
Yes, openclaw-optimize is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does openclaw-optimize support?
openclaw-optimize is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created openclaw-optimize?
It is built and maintained by Ache (@zhmza); the current version is v2.0.0.
More Skills