ClawHub
Back to skill

Security audit

中国裁判文书类案检索报告（得理科技）

Security checks across malware telemetry and agentic risk

Overview

The skill is coherent for legal case retrieval, but it should be reviewed because it can send sensitive case details to an external CLI/API and stores an API key locally with limited safety guidance.

Install only if you are comfortable using Deli Legal's CLI/API for case retrieval. Before using it with real matters, redact unnecessary personal, commercial, or privileged details, confirm when the agent is about to call the CLI, and treat the deli API key as a sensitive credential stored on the local machine.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The passive trigger condition is broad enough to auto-activate on ordinary legal discussion about how courts rule, similar cases, or supporting judgments, even when the user did not clearly request backend retrieval. In this skill, activation can lead to fact collection and deli-cli/API use, so over-triggering increases the chance of processing sensitive case information or invoking external tools without sufficiently clear user intent.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill states that if users provide case materials, the system should extract facts and use CLI-returned long-text or similar-case parameters, but it does not clearly warn users up front that uploaded legal materials may be processed through backend retrieval workflows. Because legal case files often contain highly sensitive personal, commercial, or litigation data, lack of notice and consent can create significant privacy, confidentiality, and compliance risk.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill instructs the agent to initialize the CLI by passing the user's API key on the command line and storing it in a local config file under ~/.deli/cli/config.json, but it does not warn about the security implications. Command-line secrets may be exposed through shell history, process inspection, logs, or shared workstation access, and on-disk storage increases the chance of credential theft if file permissions or endpoint security are weak.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.