Back to skill

Security audit

Openclaw Office Hours

Security checks for vulnerabilities and agentic risk

Overview

This skill is a product-planning assistant that reads project context and may create a design document, with no evidence of hidden code, credential use, or exfiltration.

Before installing, expect the skill to inspect project context and potentially create a design document in your project. Use it in repositories where that is acceptable, and specify the desired document path or ask it to show the draft before writing if you want tighter control.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs the agent to write a design document into the project directory without requiring explicit user consent at the point of modification. This is dangerous because it can cause unexpected workspace changes, overwrite user files, or create artifacts in sensitive repositories, especially when the skill is framed primarily as consultation rather than file-modifying automation.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The manifest describes an Office Hours-style consulting skill focused on reframing the problem before coding, which suggests conversational analysis. However, the documented workflow explicitly tells the agent to read repository documents, inspect git history, map code areas, and later write a design document into the project directory, expanding beyond pure consultation into repo analysis and artifact creation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.