The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run shell commands including pnpm dev, nohup, and lsof to manage the local development application and its CDP port.
[PROMPT_INJECTION]: The skill processes data from a browser context, which creates a potential surface for indirect prompt injection.
Ingestion points: Browser content is read via browser_snapshot and browser_eval as described in SKILL.md.
Boundary markers: There are no explicit delimiters or safety instructions provided to separate untrusted data from the agent's context.
Capability inventory: The skill allows the agent to execute shell commands and JavaScript within the browser.
Sanitization: No content filtering or validation is performed on data retrieved from the browser.

browser-automation