puppeteer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (LOW): The skill installs the
puppeteerpackage via npm. The source repository is maintained by Google, a trusted organization, which downgrades the risk of the download itself per the [TRUST-SCOPE-RULE]. - [COMMAND_EXECUTION] (LOW): The tool launches a Chromium browser process as a subprocess to perform automation tasks. This is the primary and expected behavior for this utility.
- [PROMPT_INJECTION] (LOW): Identified a surface for Indirect Prompt Injection (Category 8) as follows:
- Ingestion points: The skill navigates to external URLs (
page.goto) and processes arbitrary HTML strings (page.setContent), which are untrusted data sources. - Boundary markers: Absent. No delimiters or instructions are provided to the agent to ignore instructions embedded within the fetched web content.
- Capability inventory: The skill can write files to the disk (PDFs and screenshots), navigate the network, and execute JavaScript in the browser context via
page.evaluate. - Sanitization: Absent. There is no evidence of content sanitization or filtering before the agent processes or returns the data from the page.
Audit Metadata