Back to skill

Security audit

课后记录助手

Security checks for vulnerabilities and agentic risk

Overview

The skill received a suspicious verdict because a schema may allow changes to lesson-credit balances beyond the declared write scope; this is reported as a scope-enforcement gap, not evidence of malicious intent.

Consider the suspicious verdict before installing. The main reported concern is that schema validation may accept lesson-credit balance changes that should require separate authorization and teacher confirmation. The scanner inspected all 11 listed components, but analysis completeness was marked partial because two local references were unresolved, so the evidence is incomplete.

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)

The schema retains ledger balance fields despite explicitly claiming that only declared subfields remain.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

The declared write scope permits only coursePackageLedger.pendingConfirmations, but the coursePackage definition also accepts and requires totalUnits, usedUnits, and remainingUnits. Because x-skill-scope is descriptive metadata rather than an enforced JSON Schema constraint, schema validation alone would accept unauthorized balance changes, potentially bypassing teacher confirmation and corrupting lesson-credit accounting. This is a scope-enforcement gap, not evidence of malicious intent; exploitability depends on whether the runtime independently enforces field-level permissions.

Content

No source excerpt is available for this finding.