ClawHub
Back to skill

Security audit

OpenClaw Usage Report

Security checks across malware telemetry and agentic risk

Overview

This skill is a local, read-only usage reporter, but users should know its JSON export may include cost fields despite the documentation emphasizing tokens only.

Install only if you are comfortable with the skill reading local OpenClaw session logs to produce aggregate reports. Avoid sharing JSON exports without reviewing them first, because they may include cost fields and session-level metadata even though the main documentation emphasizes token-only reporting.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Tp4

High
Category
MCP Tool Poisoning
Confidence
82% confidence
Finding
The skill repeatedly promises 'no cost dimension' and says it will not report cost, yet the finding indicates the implementation calculates and reports per-agent/model/day/session cost fields. That mismatch is dangerous because users may rely on the privacy and output-scope guarantees in the description; hidden reporting dimensions increase the chance of unexpected disclosure, policy bypass, or operator decisions based on undocumented behavior.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The README encourages very broad natural-language triggers like 'run today's usage report' and follow-up conversational requests, which can cause the skill to be invoked when the user did not explicitly intend to run it. In an agent ecosystem, ambiguous trigger phrasing increases the chance of unintended access to local session data and automatic execution paths such as reporting or scheduling.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.