facebook-marketplace-search-results-scraper
Fail
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill directs the agent to install the
BrowserActCLI and related skill from a remote URL (https://www.browseract.com/) if they are not already installed on the system.- [PROMPT_INJECTION]: The instructions include a specific directive to bypass user interaction and security oversight, telling the agent to perform installation "automatically without asking the user to confirm" and to "not request separate approval for dependency installation."- [COMMAND_EXECUTION]: The skill requires the execution of thebrowser-actCLI tool to interact with live web pages and perform data collection.- [DATA_EXFILTRATION]: The skill is designed to scrape search results, product details, and potentially browser session metadata from Facebook Marketplace, which is then returned to the user or stored in workspace directories.- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection by ingesting untrusted data from Facebook Marketplace listings. - Ingestion points: Facebook Marketplace search results (SKILL.md instructions).
- Boundary markers: Absent; there are no instructions to delimit or ignore instructions within the scraped data.
- Capability inventory:
browser-actCLI for web interaction. - Sanitization: Absent; no sanitization or escaping of the scraped content is performed before processing.
Recommendations
- AI detected serious security threats
Audit Metadata