ClawHub
Back to skill

Security audit

【官方】知识星球·主题管理工具

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed ZSXQ topic-management helper that can read and change community content, including deletion, but its sensitive actions are mostly purpose-aligned and documented with confirmation guidance.

Install only if you intend to let Codex operate your logged-in ZSXQ account through zsxq-cli. Before any create, edit, reply, answer, delete, tag, or digest action, verify the exact group/topic and content; deletion is irreversible, and tag or digest changes may affect how content appears to community members.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The document explicitly notes that create, edit, reply, and answer operations require confirmation, but the delete operation is presented only as a raw DELETE command with '不可恢复' and no comparable instruction to obtain explicit confirmation. In an agent setting, this inconsistency weakens safeguards around the most destructive action and increases the chance of accidental or unauthorized permanent deletion.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill description covers a wide range of read and write operations such as searching, posting, editing, commenting, answering, deleting, and managing tags/digests, but it does not define clear activation boundaries or stronger gating for destructive/admin actions. An agent could invoke this skill in ambiguous situations and perform higher-risk actions than the user intended, especially because the same skill bundles harmless lookup actions with irreversible or privileged mutations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.