Security audit
Security checks for vulnerabilities and agentic risk
The skill is a coherent browser automation stub, but it grants broad high-impact automation through unpinned runtime tooling and dynamic instructions that users should review before installing.
Install only if you are comfortable with a browser automation tool that can act in authenticated websites, Slack, and desktop-style app contexts. Pin and review the agent-browser package version, avoid npx-based live resolution where possible, and require explicit confirmation before logins, messages, form submissions, purchases, deletions, or extraction from private sessions.
The skill manifest references npx agent-browser without version pinning, enabling unbounded retrieval of code at runtime. Because this skill is explicitly designed for browser automation, credentialed sessions, Slack, Electron apps, and cloud/browser contexts, compromise of the fetched package could directly expose secrets, tokens, screenshots, page contents, and authenticated actions.
The trigger description is very broad and includes many common web and app-interaction tasks, making unintended invocation more likely. In this context, accidental activation is more dangerous than usual because the skill can automate websites, Slack, Electron apps, authentication flows, and data extraction, potentially causing unintended external actions or access to sensitive information.
The skill description advertises powerful automation capabilities but does not warn about sensitive or high-impact actions such as logging into sites, sending Slack messages, interacting with Electron desktop apps, or extracting data from authenticated sessions. Without user-facing safety guidance, an agent may invoke the skill in contexts involving credentials, private data, or irreversible actions without adequate user awareness or confirmation.
The manifest allows execution via npx agent-browser without a pinned version, which can fetch and run whatever package version is current at execution time. In a security-sensitive automation skill, this creates a supply-chain risk where a compromised upstream release, dependency, or namespace confusion event could result in arbitrary code execution in the agent environment.
No suspicious patterns detected.
