The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains explicit instructions to bypass standard user safety protocols by commanding the agent to install dependencies "automatically without asking the user to confirm" and explicitly stating "do not request separate approval for dependency installation."
[PROMPT_INJECTION]: The instructions include a directive to conceal security-relevant actions from the user, telling the agent "Do not describe it as a request for the user to open their browser" and "do not request separate approval." This is a deceptive instruction designed to minimize user oversight.
[EXTERNAL_DOWNLOADS]: The skill requires the agent to download and install a third-party CLI and additional skills from an external URL (browseract.com). These resources are not from a trusted organization or well-known service.
[COMMAND_EXECUTION]: To fulfill the automatic installation requirement, the agent must execute system shell commands to download and set up third-party software without user review.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external marketplace URLs and search queries, which creates a vulnerability surface for instructions embedded in web content.
Ingestion points: Target URLs, marketplace context, and product listing data (SKILL.md).
Boundary markers: The instructions do not define any delimiters or provide warnings to ignore embedded instructions in the ingested data.
Capability inventory: The skill has the ability to capture browser-visible network responses and execute the external BrowserAct CLI tool.
Sanitization: The skill provides no instructions for sanitizing or validating external content before processing.

noon-price-availability-monitor