Back to skill

Security audit

数据库管理(免费版)

Security checks for vulnerabilities and agentic risk

Overview

This database helper is mostly transparent, but it asks for broad execution/write authority while teaching persistent database changes without clear safeguards.

Review before installing. Use this skill only when you want an agent to help with database-changing SQL, and require explicit approval before any CREATE, INSERT, UPDATE, DELETE, DDL, or transaction execution against a real database. Prefer test databases, verified connection targets, backups, and least-privilege credentials.

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
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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill advertises a very broad database-admin scope without clearly constraining when it should be used or requiring explicit user confirmation before suggesting or performing state-changing database actions. In an agent environment with read/exec/write tools, this can cause over-activation and lead to unsafe generation or execution of SQL against the wrong target or in inappropriate contexts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The markdown teaches table creation, inserts, and transaction execution but does not warn that these operations modify persistent user data and may be irreversible or harmful if run in production. Because the skill exposes exec capability and presents actionable SQL snippets, an agent or user could apply them directly without backup, sandboxing, or confirmation, causing unintended data changes.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.