ClawHub
Back to skill

Security audit

数据库安全审计

Security checks for vulnerabilities and agentic risk

Overview

This database security skill is purpose-aligned, but it can broadly inspect sensitive database contents, users, permissions, logs, and weak-password indicators without clear confirmation or privacy safeguards.

Install only if you intend to let the agent run database security-audit commands against databases you administer. Before use, confirm the database name and scope, prefer table-scoped scans, and treat outputs about sensitive fields, users, permissions, login IPs, audit logs, and weak passwords as confidential.

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 (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are broad, generic security-related requests such as '安全检查' and '检测注入', which can cause the skill to activate in situations where the user did not clearly intend database security operations. In this context, unintended activation can lead to unnecessary access to database metadata or sensitive outputs, especially because several mapped commands perform auditing and data inspection.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs the agent to use the skill whenever listed keywords appear, but provides no boundaries, exclusions, or confirmation requirements. This increases the chance of the agent launching security-audit commands on ambiguous requests, which is risky because the skill includes functions that inspect permissions, logs, weak passwords, and sensitive data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The sensitive-data scan is described as scanning all tables and identifying items like IDs, phone numbers, and email addresses, but it does not warn about privacy impact or limit how results are surfaced. In a database security skill, this can expose discovered sensitive fields or sample contents to the AI or end user without explicit consent, increasing confidentiality and compliance risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The weak-password check is presented without warning that it may inspect credential-related data and output a list of affected users. In practice, such functionality can reveal highly sensitive account-security information and create a targeting list for attackers or unauthorized operators if invoked casually or on the wrong audience.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.