The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the chrome-devtools-mcp package from the npm registry. This package is a vendor-attributed resource corresponding to the skill's author.
[COMMAND_EXECUTION]: The troubleshooting and setup instructions involve executing shell commands such as pgrep, lsof, pkill -9, rm, and open. These are used to manage the Chrome browser process and reset its configuration files (specifically DevToolsActivePort) to ensure a successful connection.
[REMOTE_CODE_EXECUTION]: The skill provides an evaluate_script tool that allows the agent to execute arbitrary JavaScript code within the context of the controlled browser instance.
[PROMPT_INJECTION]: The skill interacts with untrusted external web content, creating an indirect prompt injection surface.
Ingestion points: Data ingested from web pages via take_snapshot and DOM evaluation.
Boundary markers: None identified; there are no instructions to the agent to treat page content as untrusted data.
Capability inventory: The skill possesses capabilities for process management (shell commands), file system interaction (reading/deleting Chrome config files), and browser automation.
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from web pages before it is returned to the agent context.

chrome-devtools-mcp