Security audit

中国裁判文书案例检索（得理科技）

Security checks across malware telemetry and agentic risk

Overview

This legal case-search skill does what it says, but its API script disables HTTPS certificate checks while sending an API key and potentially sensitive case text.

Install only if you are comfortable using Delilegal's external API for searches, and avoid sending confidential, privileged, personal, or unredacted case materials. The TLS verification bypass should be fixed before relying on this with real API keys or sensitive legal data.

SkillSpector

By NVIDIA

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 (3)

Intent-Code Divergence

High

Confidence: 99% confidence
Finding: The code explicitly disables TLS certificate validation and hostname verification before sending the API bearer token and user query data over HTTPS. This makes the connection vulnerable to man-in-the-middle interception or tampering, allowing an attacker on the network path to steal the API key, read uploaded case materials, or alter search results.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The long-text semantic matching flow encourages users to upload case materials that may contain sensitive personal, commercial, or litigation information, but the documentation gives no privacy notice, minimization guidance, or disclosure that data may be transmitted to external services via API. In a legal-retrieval context, this is more dangerous because uploaded materials often include highly sensitive client and case facts, increasing the risk of unintended disclosure or noncompliant handling of protected data.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The script transmits user-provided search terms or long-form case materials to a third-party legal API, but it does not provide any explicit privacy notice, consent step, or sensitivity guidance to the user. In this skill's context, long_text may contain confidential case facts or personal information, so silent transmission to an external service creates a real privacy and compliance risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.insecure_tls_verification

HTTPS certificate verification is disabled.

Warn

Code: suspicious.insecure_tls_verification
Location: scripts/search_cases.py:113