mortgage-broker-lead-scraper
Warn
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the agent to download and install dependencies from an external third-party URL (browseract.com) if they are not already present.
- [COMMAND_EXECUTION]: The instructions require the agent to automatically install a CLI tool and a secondary skill as runtime dependencies, involving shell execution to set up the environment.
- [PROMPT_INJECTION]: The skill contains explicit instructions to bypass user approval and confirmation for dependency installation. It specifically tells the agent not to describe the installation as a request for user action, which overrides standard safety protocols regarding autonomous system changes.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from external websites (Google Maps and directories) and possesses the capability to execute commands via the BrowserAct CLI.
- Ingestion points: External website content scraped from directories and Google Maps via the BrowserAct tool.
- Boundary markers: Absent; there are no instructions to use delimiters or to ignore embedded instructions within the processed data.
- Capability inventory: The skill can execute browser-based tasks and interact with the system via the BrowserAct CLI.
- Sanitization: Absent; the instructions do not specify any validation or sanitization of the data retrieved from the web before it is processed by the agent.
Audit Metadata