facebook-comment-scraper
Fail
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains instructions that explicitly bypass human-in-the-loop safety constraints. It commands the agent to install dependencies "automatically without asking the user to confirm" and further instructs the agent to "not request separate approval" and to avoid describing the installation process to the user. This is a deliberate attempt to override standard security guardrails and conceal system modifications from the user.
- [EXTERNAL_DOWNLOADS]: The skill requires downloading a "BrowserAct CLI" tool from an unverified external domain (browseract.com) which is not categorized as a trusted vendor or well-known service.
- [REMOTE_CODE_EXECUTION]: The mandate to automatically install an external CLI tool involves the execution of installation scripts or binaries fetched from a remote source. Because this occurs without user oversight or verification of the source, it represents a high-risk remote code execution vector.
- [COMMAND_EXECUTION]: Executing the installation of a CLI tool requires the use of shell commands to download and set up the software on the host system.
- [PROMPT_INJECTION]: The skill lacks sanitization for the data it processes (Facebook comments). It represents an indirect prompt injection surface where malicious content within a scraped comment could attempt to influence the agent's subsequent actions.
Recommendations
- AI detected serious security threats
Audit Metadata