Openclaw Performance Skill

• Outcome: Review
• Latest audit: May 29, 2026 at 9:35 PM UTC
• Version:
  1.0.0

Overview

This skill appears intended for OpenClaw diagnostics and optimization, but it can persistently read and change local OpenClaw configuration without enough user control or disclosure.

Review before installing. Use it only if you are comfortable with it reading local OpenClaw configuration and changing ~/.openclaw/config.json. Prefer running it in dry-run or review mode if available, back up your config first, and avoid using it on workspaces where reports could capture secrets or sensitive infrastructure details.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (4)

Missing User Warnings

Medium

• Confidence: 83% confidence
• Finding: The skill advertises 'one-click optimization' and monitoring deployment, which implies actions that may modify configuration files, write artifacts, or install/enable monitoring components, but it does not clearly disclose those side effects. In an agent context, vague automation descriptions can lead users to authorize changes without understanding scope, increasing the risk of unintended system modification or privacy-impacting telemetry deployment.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The script reads configuration files from the user's home directory without clearly warning the user that personal local configuration will be accessed. In a skill/agent context, undisclosed inspection of home-directory files can surprise users, expose sensitive tokens or endpoints contained in config files, and normalize overbroad local file access.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The script writes a diagnostic report to disk automatically without prior warning or confirmation. In an agent skill context, silent file creation can leave sensitive environment details, system state, and configuration-derived data on disk where other users, processes, or future tooling may access them unexpectedly.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The script writes directly to the user's ~/.openclaw/config.json after merging settings, without any confirmation, dry-run mode, or explicit opt-in. In a one-click optimization skill, silently modifying persistent user configuration can cause unexpected behavior, break existing setups, or replace security-relevant settings, especially because the script also creates the target directory and proceeds even when prior config parsing fails.

VirusTotal

66/66 vendors flagged this skill as clean.