Security audit

Legal Ai Counsel

Security checks for vulnerabilities and agentic risk

Overview

This is a China-focused legal information helper with disclosed limits and no evidence of hidden access, persistence, exfiltration, or destructive behavior.

Install this only if you want Chinese-language, China-oriented legal information and preliminary document or cost assistance. Do not rely on it as a lawyer, especially for urgent, jurisdiction-specific, criminal, family, immigration, or high-value matters; consult a qualified attorney for decisions or filings.

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
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 (4)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The invocation description uses broad phrases such as '法律帮助' and '法律建议', which can overlap with ordinary requests for general help and cause the skill to trigger in situations where the user did not explicitly request legal assistance. In a legal context, over-triggering is more sensitive because users may receive legal-style guidance or disclaimers in place of the intended general assistance, increasing the risk of inappropriate routing and misplaced reliance.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The skill metadata and examples strongly bias toward Chinese-language interaction without stating that the assistant should match the user's language or offer a language choice. This can cause usability and comprehension failures, which are especially problematic in legal guidance where misunderstanding terminology, rights, or procedures may materially affect user decisions.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: All user-facing messages, disclaimers, task labels, and CLI examples are in Chinese only, and the handler provides no option to choose another language or locale. This creates a natural-language policy issue because the skill effectively enforces a specific language without user opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The description states the skill is for '中国场景' and is written entirely in Chinese, indicating a fixed China-specific language/locale context. There is no indication that users can choose another language or locale, or that the restriction is presented as an explicit opt-in.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.