The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external web pages, creating a surface for instructions embedded in those pages to influence the agent's behavior.
Ingestion points: The skill reads page structure and text via mcp__claude-in-chrome__read_page, mcp__chrome-devtools__take_snapshot, and mcp__claude-in-chrome__get_page_text as documented in SKILL.md.
Boundary markers: Absent; the instructions do not specify using delimiters or warnings to ignore instructions embedded within the processed web content.
Capability inventory: The skill possesses the capability to execute JavaScript in the browser (mcp__chrome-devtools__evaluate_script), perform network operations, and write files (performance traces and GIF exports) as seen in SKILL.md. It also spawns a local subprocess in scripts/main.cjs.
Sanitization: Absent; there is no mention of filtering, escaping, or validating the content retrieved from web pages before processing.
[DYNAMIC_EXECUTION]: The skill includes tools that allow for the dynamic execution of JavaScript code within the browser context.
Evidence: SKILL.md documents the use of mcp__chrome-devtools__evaluate_script and mcp__claude-in-chrome__javascript_tool for custom checks and page interaction.
[COMMAND_EXECUTION]: The skill's primary entry point executes a local tool file via a subprocess.
Evidence: scripts/main.cjs uses child_process.spawn to execute a local JavaScript file located at .claude/tools/chrome-browser/chrome-browser.cjs.
[EXTERNAL_DOWNLOADS]: The documentation points to external installation requirements from well-known and trusted sources.
Evidence: SKILL.md references the official Google Chrome download site and the Claude-in-Chrome extension on the Chrome Web Store.

chrome-browser