Back to skill

Security audit

排课与课时管理

Security checks for vulnerabilities and agentic risk

Overview

ClawScan labels this skill benign with medium confidence: it found one concern about parent-facing replies extending beyond the stated scope, while the other reported checks found no issue; analysis was partial.

Consider whether you are comfortable with the skill’s reported instruction to compose parent-facing replies despite its stated scope. Skillspector inspected all 12 listed components, but analysis was marked partial because two local references could not be matched to bundled files; AIG reported no findings. Treat the benign verdict as a favorable assessment with incomplete evidence, not as a guarantee of safety.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

L056 instructs the skill to compose parent-facing replies, although the manifest explicitly excludes drafting parent messages and delegates parent communication to parent-communication.

Medium
Category
Excessive Agency
Confidence
88% confidence
Finding

The manifest states that this scheduling skill does not draft parent messages. L056 nevertheless directs it to add specific language when replying to a parent and supplies an alternative reply for a protective parent, extending its behavior beyond crisis referral into explicitly excluded parent communication.

Content

No source excerpt is available for this finding.