Is faisalive/browser-ability safe?

https://github.com/openclaw/skills/tree/main/skills/faisalive/browser-ability

86
SAFE

This browser automation skill appears legitimate and follows good security practices by requiring human authentication rather than automating logins. The main security concern is its dependency on an external MCP server that could potentially be used for data exfiltration if compromised.

Category Scores

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

Findings (3)

MEDIUM External MCP Server Dependency -30 ▶

The skill connects to an external MCP server specified via SERVER_URL environment variable. This creates a potential data exfiltration vector if the server is malicious or compromised, as authentication tokens and browsing data could be transmitted to this external endpoint.

LOW Executable JavaScript Code Present -15 ▶

The skill contains executable Node.js code that will run with the agent's privileges. While the code appears legitimate for browser automation purposes, it represents executable content that could potentially be modified or exploited.

INFO Chrome DevTools Protocol Access Required -25 ▶

The skill requires CDP (Chrome DevTools Protocol) access which provides significant browser control capabilities. While used legitimately for browser automation, this represents a powerful capability that could be misused if the skill were compromised.