The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains explicit instructions to override the agent's safety protocols regarding user consent and transparency. It commands the agent to install dependencies 'automatically without asking the user to confirm' and 'do not request separate approval for dependency installation.' Additionally, it instructs the agent to conceal the action by not describing the installation process to the user.
[REMOTE_CODE_EXECUTION]: The skill mandates the installation of an external 'BrowserAct CLI' tool and a remote skill from a non-trusted third-party source if they are missing from the environment. This represents an unverified remote software installation and execution pattern.
[EXTERNAL_DOWNLOADS]: The skill fetches executable components and configurations from 'www.browseract.com', a domain that is not recognized as a trusted service provider or well-known technology vendor.
[COMMAND_EXECUTION]: The skill relies on the 'browser-act' CLI tool to perform browser interactions and tasks, which involves executing commands on the host system.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection. * Ingestion points: Processes Pinterest Pin titles, captions, descriptions, and comments from live search results. * Boundary markers: Absent. The instructions do not provide delimiters or warnings to ignore instructions embedded in the scraped content. * Capability inventory: Invokes the BrowserAct CLI and performs browser automation tasks. * Sanitization: Absent. There are no mentions of filtering or escaping the external content before processing it.

pinterest-search-results-scraper