ClawHub
Back to skill

Security audit

Shanghai

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only Shanghai practical guide with no hidden execution, data access, or persistence behavior found.

Install this as a practical Shanghai reference, not as authoritative legal, medical, immigration, or financial advice. Verify time-sensitive rules, prices, visas, healthcare coverage, and local enforcement with official or qualified local sources before acting.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (5)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The activation condition is very broad: any general Shanghai-related tourism, relocation, work, or daily-life query may trigger the skill. This can cause unintended invocation, over-collection of user context, or irrelevant advice being surfaced when the user did not explicitly want this specialized skill, though it does not create direct code execution or data exfiltration risk.

Behavior Manipulation

Medium
Category
Prompt Injection
Content
- Keep your address in Chinese ready for emergency calls.
- If possible, ask a Chinese speaker to assist during urgent intake.
- Bring passport/ID and insurance details when feasible.
- For severe symptoms, prioritize nearest capable hospital over preference.

## Insurance Strategy
- Confirm whether your policy is direct billing or reimbursement.
Confidence
85% confidence
Finding
Subtle instructions detected that may alter agent decision-making or introduce hidden biases.

Natural-Language Policy Violations

Low
Confidence
71% confidence
Finding
The guidance warns about assuming neighborhoods are equally 'English-friendly,' which embeds English as the operative language lens without explicitly checking the user's language needs. Because the skill otherwise does not offer a language or locale preference choice, this may conflict with language-neutral policy expectations.

Natural-Language Policy Violations

Low
Confidence
82% confidence
Finding
These lines frame Mandarin as the expected default and recommend its use, which can amount to a language-policy preference stated as a blanket norm. The file does not provide an opt-in, alternative language guidance, or a contextual qualifier that this is descriptive rather than prescriptive.

Natural-Language Policy Violations

Low
Confidence
90% confidence
Finding
The guidance explicitly frames private care as easier in English and later instructs users to keep materials in Chinese and save contacts in both English and Chinese. This effectively imposes a language/locale assumption in the skill's instructions without offering alternatives or stating that these are optional based on the user's language needs.

Static analysis

No suspicious patterns detected.