ClawHub
Back to skill

Security audit

数学测评设计

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only math assessment design skill whose student-data integrations are disclosed and purpose-aligned, though users should confirm before any student record write-back.

Installers should treat this as an education-data workflow: use pseudonyms where possible, confirm before reading student profiles or writing results to downstream systems, and avoid producing rankings or public score disclosures unless the teacher explicitly authorizes an appropriate use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The activation phrases are broad and overlap with ordinary teacher discussion, which can cause the skill to trigger in contexts where the user did not intend assessment-design behavior. Unintended invocation can lead to inappropriate data flow into connected education-analysis skills or generation of outputs that affect student records and teaching decisions without clear user intent.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger section lacks exclusion criteria, so the skill may activate for generic mentions of tests, difficulty, or student assessment even when the user is only asking casually or discussing policy. In a system with downstream integrations, poor activation boundaries increase the risk of over-collection, unintended persistence, or context switching into a more sensitive workflow than the user requested.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill description says assessment results may be written into downstream student-analysis systems, but it does not present a prominent user-facing warning or consent boundary at the point of capability description. Because this skill operates in an education context involving student performance data, silent or unclear propagation of results raises privacy and data-minimization concerns.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.