ClawHub
Back to skill

Security audit

工作周报技能

Security checks for vulnerabilities and agentic risk

Overview

This skill locally turns user-provided production data into weekly reports, with disclosed file reading and archiving behavior that fits its purpose.

Install only if you are comfortable with the agent reading the production files you provide and saving report copies under the weekly_reports history directory. Keep sensitive inputs and generated reports in a controlled workspace, and review output paths before running the helper scripts.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill instructs the agent to read user-supplied Word/Excel files and write/archive HTML reports, but the skill metadata declares no permissions or constraints for those capabilities. This creates a trust and containment gap: an agent may access or modify local files without explicit user-visible authorization boundaries, increasing the risk of unintended file access or persistence.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger conditions are broad enough that the skill could activate for loosely related requests such as generating reports or reviewing historical data without clear user intent to invoke file-processing behavior. Overbroad invocation increases the chance of unintended parsing, history reads, or file creation in contexts where the user did not expect those actions.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill describes generating output files and automatically archiving them, but it does not require an explicit warning or user confirmation before writing persistent artifacts. That can lead to unintended data retention, overwriting, or storage of sensitive production information in local history directories without the user's informed consent.

Static analysis

No suspicious patterns detected.