Back to skill

Security audit

Openclaw Ceo Review

Security checks for vulnerabilities and agentic risk

Overview

This skill is a strategy-review helper whose repository inspection is read-only and purpose-aligned, though users should know it asks the agent to inspect local project context.

Install this if you want a rigorous plan-review workflow that can inspect the current repository for context. Avoid using it on projects where commit history, TODOs, architecture docs, or design notes should not be read by the agent session unless you are comfortable sharing that context.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly instructs the agent to run multiple repository-inspection shell commands on the local system before doing any review, but it provides no user warning, confirmation gate, or limitation on when command execution is allowed. Even though the listed commands are read-oriented, they still access local repository contents and metadata, which can expose sensitive filenames, commit history, design docs, TODOs, and architectural information without explicit user consent.

Content

No source excerpt is available for this finding.