Security Audit

browserless-automation

github.com/ComposioHQ/awesome-claude-skills

AI SkillCommit 99e2a2951545

TRUSTED

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

browserless-automation received a trust score of 86/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.

SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Exposure of Powerful Web Automation Capabilities.

The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.

Last analyzed on February 17, 2026 (commit 99e2a295). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

85%

Behavioral Risk Signals

Network Access

1 finding

Dynamic Code

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Exposure of Powerful Web Automation Capabilities The skill exposes the LLM agent to the 'browserless' toolkit via Rube MCP, which provides extensive web automation capabilities. These capabilities include navigating to arbitrary URLs, interacting with web page elements, extracting data, and submitting forms. If the LLM agent's instructions are compromised (e.g., via prompt injection), a malicious actor could leverage these tools to perform unauthorized actions on behalf of the user, exfiltrate sensitive data from web pages, or navigate to malicious sites. The skill itself is a wrapper around these powerful tools, making the agent highly capable but also a high-value target for misuse. Implement strict input validation and sanitization for all agent prompts, especially when interacting with tools that have broad web access. Consider sandboxing the browser environment used by the 'browserless' toolkit. Require explicit user confirmation for sensitive web actions (e.g., navigating to new domains, submitting forms, or extracting data from specific fields). Limit the scope of URLs the 'browserless' toolkit can access if possible. Educate users about the risks of providing untrusted input to agents with web automation capabilities.	LLM	SKILL.md:1

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/b91f9a52e794d9e5.svg)](https://skillshield.io/report/b91f9a52e794d9e5)