ClawHub
Back to skill

Security audit

文档数据库管理(免费版)

Security checks for vulnerabilities and agentic risk

Overview

The skill is clearly aimed at MongoDB Atlas administration, but it asks an agent to run broad live admin API actions with credentials while giving incomplete safety boundaries and packaging no referenced scripts.

Review this skill carefully before installing. Only use it with least-privileged MongoDB Atlas API credentials, restrict it to known projects, run dry-runs first, and require explicit human approval before any live `--yes` operation. The package also appears incomplete because the scripts referenced by the instructions are not included in the artifact inspected here.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill documents two different credential schemes: Atlas-specific variables (`ATLAS_CLIENT_ID` / `ATLAS_CLIENT_SECRET`) and a generic `API_KEY` export. This inconsistency can cause users or agents to place secrets in the wrong variable, break authentication flows, or accidentally expose unrelated credentials during troubleshooting or execution.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly documents live administrative API execution with `--yes` automatic confirmation for operations like cluster creation, but does not prominently warn that these actions can make real, potentially costly or disruptive changes to a MongoDB Atlas environment. In an agent context, this increases the risk of unintended infrastructure modification because users may treat the examples as routine rather than high-impact admin actions.

Missing User Warnings

Low
Confidence
84% confidence
Finding
The skill instructs users to place sensitive Atlas credentials in environment variables in the main workflow without an immediate warning about secret handling, shell history, process inheritance, or accidental logging by tools and agents. While environment variables are common, omission of handling guidance raises the chance of credential leakage in multi-tool automated environments.

Static analysis

No suspicious patterns detected.