vibebrowser
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill dynamically downloads and executes the latest version of the
@vibebrowser/clipackage from the NPM registry usingnpx. This is an external dependency managed by the skill's author (vibetechnologies). - [COMMAND_EXECUTION]: The skill makes extensive use of shell commands via
npxto interact with the Vibe Browser CLI. It passes parameters such as relay URLs and page IDs derived from previous command outputs or user-provided values. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process arbitrary content from the user's browser (e.g., page snapshots, tab titles, and JavaScript evaluation results). This creates a surface where malicious content on a webpage could attempt to influence the agent's logic or behavior.
- Ingestion points: Browser data is retrieved through
snapshot,tabs, andevaluatecommands as specified inSKILL.md. - Boundary markers: None identified; instructions direct the agent to read and act upon the raw browser content directly.
- Capability inventory: Subprocess execution via
npxand browser-side JavaScript execution via theevaluatecommand (referenced inSKILL.md). - Sanitization: There are no instructions for sanitizing, escaping, or filtering the data retrieved from the browser before it is processed by the agent.
Audit Metadata