Back to skill

Security audit

Huo15 Openclaw Security Review

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed security-review checklist that scans code changes and only fixes issues after user approval.

Before installing, understand that this skill may read code diffs and relevant file context to find security issues, including possible secrets. Invoke it with an explicit branch, file list, or commit range when scope matters, and review any proposed fixes before approving changes.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Cloud Metadata Access

High
Category
Server-Side Request Forgery
Confidence
90% confidence
Finding

Code accesses a cloud instance metadata endpoint (e.g. 169.254.169.254). A single request can return temporary IAM credentials, making this a high-value SSRF target for credential theft.

Content

Scanner excerpt · SKILL.md (reported line 73)May include surrounding context.

md
| 检查项 | 信号 |
|--------|------|
| `fetch(userURL)` 无白名单 | 可被打到 169.254.169.254(云 metadata)|
| 允许 `file://` / `gopher://` 协议 | scheme 未校验 |
| DNS rebinding 窗口 | 二次解析 |

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill declares very broad aliases and trigger phrases such as '安全评审', 'security review', and '漏洞检查', which are common natural-language requests that may overlap with other review-oriented skills. In an agent environment, ambiguous invocation can cause this skill to activate unexpectedly, expanding its operational scope and potentially exposing repository contents or causing unintended security-review actions in contexts the user did not explicitly choose.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The example activation scenarios are generic conversational requests like '帮我做个 security review' and '这个 PR 有没有安全问题', without strong constraints tying them uniquely to this skill. This increases the chance of accidental or adversarial triggering in mixed-skill environments, especially because the skill can inspect diffs and sensitive code changes by default.

Content

No source excerpt is available for this finding.