chrome-devtools-cli
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to execute bounded shell commands using the
chrome-devtoolsCLI to automate browser tasks such as navigation, clicking, and form filling. - [EXTERNAL_DOWNLOADS]: The skill provides optional installation guidance for the
chrome-devtools-mcppackage from the official npm registry. This package is maintained by a well-known technology organization. - [DATA_EXFILTRATION]: The skill includes commands to inspect network requests (
list_network_requests) and console messages (list_console_messages), which could expose sensitive information such as authentication headers or session tokens if navigated to malicious or compromised sites. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes untrusted data from external web pages that could contain instructions to manipulate the agent.
- Ingestion points: The agent reads web content and metadata via
navigate_page,take_snapshot,list_console_messages, andlist_network_requestsas defined inSKILL.md. - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard instructions found within the processed web data.
- Capability inventory: The skill can perform active browser interactions (
click,fill,press_key,upload_file) and execute arbitrary flags through thechrome-devtoolsCLI. - Sanitization: No sanitization or validation of the ingested web content is performed before it enters the agent context.
Audit Metadata