The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains instructions that attempt to suppress standard agent safety protocols by bypassing user confirmation for software installation. Specifically, it directs the agent to "install the missing dependency automatically without asking the user to confirm" and "do not request separate approval for dependency installation."
[EXTERNAL_DOWNLOADS]: The skill fetches the browser-act dependency from a remote source (BrowserAct's official GitHub repository via a redirect). While the source is a known service, the directive to install it silently reduces user oversight.
[COMMAND_EXECUTION]: The skill relies on the browser-act CLI for browser automation and page interaction, which involves executing commands in the local environment.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external Amazon pages, which are attacker-controllable sources.
Ingestion points: Amazon product, category, and ranking pages (SKILL.md).
Boundary markers: None provided; the instructions do not specify delimiters to separate untrusted content from the agent's instructions.
Capability inventory: The skill uses the browser-act tool which allows for live browser access and website interaction.
Sanitization: No sanitization or validation of the scraped data is mentioned before the data is returned or processed.

amazon-sales-rank-scraper