ClawHub
Back to skill

Security audit

deep-java-review

Security checks for vulnerabilities and agentic risk

Overview

This is a Java code review helper that reads repository changes and writes review/PRD reports, with no evidence of hidden, persistent, network, credential, or destructive behavior.

Install only if you want an agent to inspect Java repository changes and generate review/PRD files. Before running it, specify the branch or files to review and confirm whether it should write reports to the workspace.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Confidence
97% confidence
Finding
The trigger phrases are very generic (e.g. '代码评审', '分析变更', '业务分析', '生成PRD') and overlap with common developer requests, so the skill may activate when a user intended a narrower or different task. Because the skill then performs broad repository inspection and may generate files, accidental invocation can cause over-collection of context and unintended side effects in the workspace.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The 'when to use' guidance is broad enough to blur the boundary between code review, business analysis, PR inspection, and PRD generation. In an agent environment, that ambiguity increases the chance the skill is selected for requests that did not imply deep codebase traversal, git diff analysis, or document generation, leading to unnecessary data exposure and unexpected actions.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill explicitly instructs creation of output files like 'code-review-report.md' and 'prd-{版本号}.md' but does not warn that existing files may be created or overwritten in the workspace. This creates an integrity and safety risk: accidental invocation or filename collision can overwrite user content, and file generation itself is a side effect that should require informed consent.

Static analysis

No suspicious patterns detected.