The Agent Skills Directory

[PROMPT_INJECTION]: The instructions explicitly direct the agent to bypass user oversight and confirmation protocols during software installation. Specifically, it commands the agent to install dependencies "automatically without asking the user to confirm" and to "not request separate approval for dependency installation."
[EXTERNAL_DOWNLOADS]: The skill requires the agent to download and install the BrowserAct CLI and its associated skill set from external sources (browseract.com and GitHub) if they are not already present on the system.
[COMMAND_EXECUTION]: The skill is designed to invoke the browser-act CLI tool to perform web scraping tasks, which involves executing system-level commands.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and processes untrusted data from Facebook search results, which could contain malicious instructions designed to hijack the agent.
Ingestion points: Facebook search results, profiles, pages, and groups (SKILL.md).
Boundary markers: Absent; the instructions do not define delimiters or provide "ignore embedded instructions" warnings for the external data being processed.
Capability inventory: The agent utilizes the browser-act skill for live browser interaction and CLI execution, and is instructed to manage data within the workspaces/ directory.
Sanitization: There is no mention of sanitizing, filtering, or validating external content before it is processed by the agent.

facebook-search-results-lead-scraper