ClawHub
Back to skill

Security audit

Minimax Usage Cn

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Minimax quota checker that uses an API key to read usage data, with no evidence of hidden exfiltration, destructive actions, or automatic persistence.

Before installing, confirm you trust www.minimaxi.com with the MINIMAX_API_KEY and only run the script for quota checks. If you use the cron examples, adapt the hardcoded paths, ensure the API key is available to cron intentionally, choose a safe log location, and document how to remove the scheduled job.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill explicitly relies on shell execution (`curl`, shell script, cron examples) but does not declare any tool scope or allowed-tools constraints. That creates an unnecessary trust gap: an agent may execute shell-capable steps without an explicit least-privilege boundary, increasing the chance of unintended command execution or broader system access than the skill actually needs.

Session Persistence

Medium
Category
Rogue Agent
Content
```bash
# 编辑 crontab
crontab -e

# 添加以下行（每小时检查并记录）
0 * * * * cd /home/rocfly/.openclaw/workspace/skills/minimax-usage-cn/scripts && ./minimax-usage.sh >> /home/rocfly/.openclaw/workspace/logs/minimax-usage.log 2>&1
Confidence
85% confidence
Finding
The skill recommends installing a recurring cron job, which creates persistence outside the immediate user session. Even though the stated purpose is quota monitoring, persistence mechanisms are security-relevant because they continue executing unattended, can expose API-derived data to logs, and may normalize long-lived scheduled execution without explicit lifecycle controls.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The script labels fields as 'remains/remaining' but actually parses `current_interval_usage_count` and `current_weekly_usage_count`, which appear to be usage counters rather than remaining quota. This can invert the operator's understanding of quota state, causing users or automation to believe capacity is available when limits are nearly exhausted, or vice versa.

Natural-Language Policy Violations

Low
Confidence
88% confidence
Finding
The notes state that the window time is in UTC+8, and the skill title/description also indicate a China-specific variant. While region-specific tooling can justify locale constraints, this file does not clearly present the timezone choice as optional or explain it as a user-selectable display preference.

Static analysis

No suspicious patterns detected.