Back to skill

Security audit

物理实验教学指导

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because its schema may accept or expose fields beyond its stated data scope; the scan is partial, so the evidence is incomplete.

Review the schema's accepted fields and confirm that access controls enforce the intended data scope before installing. Treat the verdict as a caution, not a claim of malicious intent; some analysis was partial because two referenced paths were missing.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

The schema contradicts its claim that all subfields were trimmed to the skill's declared scope.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The schema claims its subfields are restricted to declared access scope, but classProfile retains undeclared fields such as classId, subject, and gradeBand, with several required; other definitions likewise retain fields outside their declared lists. If a consumer relies on this schema as the scoped-data filter, those extra fields can be accepted or exposed despite the stated boundary, although the supplied file alone does not establish a runtime authorization bypass.

Content

No source excerpt is available for this finding.