ClawHub
Back to skill

Security audit

中国邮政储蓄银行银企直联官网

Security checks for vulnerabilities and agentic risk

Overview

This skill only directs the agent to output the official PSBC website URL, with no code, credentials, persistence, or account-changing actions.

Before installing, treat this as a simple PSBC website locator rather than a banking API or transaction skill. The publisher should narrow the description and add clear trigger conditions, but the current artifact does not ask for credentials, modify accounts, run code, or persist anything.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (3)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manifest description at L03-L04 claims the skill supports comprehensive corporate banking services such as account queries, fund transactions, receipts, payroll, bills, cross-border finance, and digital RMB. However, the actual documented behavior in L13-L20 is limited to visiting the PSBC official site and returning its URL, which is far narrower and does not implement the described banking-service capabilities.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

Natural-language policy checks apply to all file types. The file consistently specifies Chinese-language content and output, but does not offer language choice or explain that the skill is intentionally limited to a Chinese-only context, which may violate language/locale policy expectations.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This is a markdown file, so vague-trigger checks apply. The 'When to Run' section is present but empty, leaving it unclear what user requests should activate this skill and what requests should not, which can cause unintended invocation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.