browser-testing-with-devtools
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: Static analysis hints regarding instruction overrides are false positives. The skill text mentions patterns like 'Ignore previous instructions' specifically to teach the agent to disregard such commands if encountered within untrusted browser content.
- [EXTERNAL_DOWNLOADS]: The skill configuration uses 'npx' to fetch and run 'chrome-devtools-mcp'. This is the standard method for deploying this tool and is consistent with the skill's stated purpose of browser automation and testing.
- [COMMAND_EXECUTION]: The skill facilitates the execution of the Chrome DevTools MCP server. It implements security best practices by recommending the '--isolated' flag to ensure that testing occurs in a temporary browser environment, preventing access to the user's main profile.
- [DATA_EXFILTRATION]: The skill processes untrusted data from web pages (DOM, logs, network traffic). It mitigates exfiltration risks through explicit rules forbidding the agent from accessing cookies, session tokens, or local storage, and by requiring user confirmation before navigating to any URL found on a page.
Audit Metadata