The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the use of the Lighthouse performance auditing tool via npx for automated performance measurements.
[PROMPT_INJECTION]: The skill defines workflows where an agent interacts with live browser runtime data, creating a surface for indirect prompt injection attacks.
Ingestion points: The skill utilizes Chrome DevTools MCP to inspect external web pages, including DOM state and network requests (SKILL.md).
Boundary markers: No delimiters or safety instructions are provided to help the agent distinguish between legitimate tool output and malicious instructions potentially embedded in web content.
Capability inventory: The skill provides access to browser inspection, performance measurement, and network log data (SKILL.md).
Sanitization: There are no mechanisms described for sanitizing data ingested from the browser environment before it is processed by the agent.

browser-devtools