Back to skill

Security audit

Cmdnotify

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent command-monitoring guide, but users should be careful because its optional alert commands can send or store monitored output.

Install only if you want a tool that repeatedly runs local commands you configure. Review every monitored command and notify_cmd before use, avoid sending secrets or full command output to webhooks, and avoid persistent logs that may retain sensitive alert content.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill documentation is written entirely in Chinese, and there is no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy, forcing a specific language without opt-in is a natural-language locale violation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly documents using a custom notification command to send alert content to external services, and the example includes posting $CMD_MESSAGE to a remote Slack webhook. Because monitored command output may contain sensitive system details, credentials, URLs, or operational metadata, this creates a real data-exposure risk when used without prominent warning, scoping, or redaction guidance.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

The documented curl-based notify_cmd sends alert content to an external webhook, which is a concrete example of external transmission. In the context of a command-monitoring skill, alert payloads can contain command output, error text, host health details, or other sensitive information, so encouraging this pattern without safeguards materially increases the risk of data leakage.

Content

Scanner excerpt · SKILL.md (reported line 114)May include surrounding context.

json
{
  "notify_cmd": "curl -X POST -d '{\"text\":\"$CMD_MESSAGE\"}' https://hooks.slack.com/services/..."
}

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The Advanced Example uses notify_cmd to append alert messages to /var/log/alerts.log, which affects local system state by writing persistent data. The skill description does not call out this file-writing behavior or caution users about log retention and sensitive content in alerts.

Content

No source excerpt is available for this finding.