ClawHub
Back to skill

Security audit

SQL查询工具(免费版)

Security checks for vulnerabilities and agentic risk

Overview

This SQL skill is coherent but should be reviewed because it enables broad database write and delete workflows without clear user-confirmation guardrails.

Install only if you want an agent to help operate databases through command-line tools. Prefer read-only database credentials by default, require explicit confirmation before any write, delete, schema, or production operation, and review generated SQL before execution.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger conditions are broad enough to match many ordinary database-related requests, which can cause an agent to invoke a write-capable SQL skill in contexts where the user did not explicitly request execution. Because the skill advertises exec/write tooling and supports modification/deletion flows, overbroad activation increases the chance of unintended database access or destructive operations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill exposes write-capable and potentially destructive database operations but does not present a clear warning near the usage instructions that these actions may alter or delete production data. In an agent setting, this omission is dangerous because users may assume the tool is primarily analytical while the documented modes include create/modify/delete and transaction-wrapped writes.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest emphasizes SQL query execution, parameterized access, EXPLAIN analysis, and portability, which reads primarily like a querying/tuning tool. However, L047 explicitly states support for create/query/update/delete modes, expanding behavior into full data mutation beyond the narrower query-analysis framing.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
67% confidence
Finding

The manifest repeatedly frames the free edition around cross-database portability and migration basics, and L068 lists migration-related capability coverage. Later, L207 says the free version does not support cross-database SQL automatic conversion, creating a semantic mismatch about how much migration assistance the free edition actually provides.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.