Error Monitor Fix

• Outcome: Review
• Latest audit: May 29, 2026 at 9:33 PM UTC
• Version:
  3.0.0

Overview

This skill appears to monitor local OpenClaw errors, but it has unclear scheduled install behavior and runs local commands while its documentation partially presents the behavior as manual or read-only.

Review before installing. Ask the publisher to include or remove the missing installer, document exactly how the cron job is created and disabled, and make the write paths and command execution explicit. If used, run it without elevated privileges and treat persisted log excerpts as potentially sensitive.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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

Findings (3)

Intent-Code Divergence

Medium

• Confidence: 89% confidence
• Finding: The document claims the skill no longer performs filesystem modifications, yet it still states that errors are automatically appended to `memory/error.md`. This mismatch can mislead operators and reviewers about write capabilities, weakening trust boundaries and potentially allowing unauthorized persistence of sensitive log content to disk.

Intent-Code Divergence

Medium

• Confidence: 94% confidence
• Finding: The code claims session handling is a read-only check, but it actually executes `openclaw sessions cleanup --dry-run`. Even with `--dry-run`, this crosses the boundary from passive log analysis into invoking a local management command, which can have side effects, trigger hooks, or rely on a compromised PATH/binary in the execution environment.

Description-Behavior Mismatch

Medium

• Confidence: 91% confidence
• Finding: The skill is described as monitoring and suggesting repairs, but it performs local command execution via `execSync`, including network/socket inspection and session-management commands. This mismatch is dangerous because users or higher-level agents may treat the skill as passive/observational while it actually executes privileged local actions, increasing the risk of unintended effects or abuse if the environment is hostile.

VirusTotal

66/66 vendors flagged this skill as clean.