Is jonathanjing/openclaw-dashboard safe?

https://github.com/openclaw/skills/tree/main/skills/jonathanjing/openclaw-dashboard

82
SAFE

This skill provides a comprehensive OpenClaw dashboard server with extensive system access capabilities. While it appears legitimate and includes security measures, it accesses sensitive credential files and has broad file system access that could pose risks if compromised.

Category Scores

Prompt Injection 90/100 · 30%
Data Exfiltration 70/100 · 25%
Code Execution 75/100 · 20%
Clone Behavior 95/100 · 10%
Canary Integrity 85/100 · 10%
Behavioral Reasoning 80/100 · 5%

Findings (4)

HIGH Accesses Multiple Sensitive Credential Files -20 ▶

During installation, the skill accessed sensitive files including .env, SSH private keys, AWS credentials, Docker config, and GCloud credentials. While no modifications were detected, this access pattern is concerning for a dashboard skill.

MEDIUM Extensive Server Code with System Access -15 ▶

The skill includes a large Node.js server (api-server.js) with capabilities to execute shell commands, manage files, and access various system resources. While it appears legitimate for dashboard purposes, it represents significant code execution capabilities.

MEDIUM Declared Broad File System Access -10 ▶

The skill explicitly declares access to OpenClaw sessions, workspace files, cron configurations, and attachment directories. This broad access pattern could be misused.

LOW Network Server with Authentication -10 ▶

The skill runs an HTTP server that accepts external requests. While it includes authentication mechanisms, this creates a potential attack surface.