ClawHub
Back to skill

Security audit

专业行业研究报告生成器

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed industry-report generator, with some over-broad workflow behavior users should be aware of before use.

Install this if you want a full report-generation workflow. Expect it to perform web research, query in both Chinese and English, and leave Markdown, JSON, DOCX, and PDF files in the workspace; do not invoke it with sensitive topics unless that exposure and persistence are acceptable.

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
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger conditions are broad enough to activate on many generic requests about analysis or understanding an industry, which can cause the skill to run unexpectedly and take over interactions that the user may not have intended to turn into a full report-generation workflow. In this skill, that overreach is more dangerous because activation leads to multi-step autonomous research and file generation, increasing the chance of unnecessary external searches, excessive workspace writes, and user-confusing behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

Mandating both Chinese and English searches without user choice can cause unnecessary expansion of data collection scope, including sending user topics to additional search contexts and jurisdictions that may not be relevant. In a research skill, this increases privacy, compliance, and relevance risks because the agent may query more broadly than needed and incorporate locale-mismatched sources into the final report.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill instructs creation of multiple intermediate files plus final DOCX/PDF outputs without requiring any explicit notice or consent to the user. This can lead to unexpected persistent artifacts in the workspace, possible storage clutter, and accidental handling of sensitive user-provided content in generated files, especially since the workflow is mandatory and not conversational.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.