google-dorks-catalog
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill defines network operations to non-whitelisted external domains, such as using
curlto verify exposed files ontarget.comand querying the GitHub Search API. These operations are used for reconnaissance but represent an external communication surface. - [PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted content from search engines and code repositories. This behavior creates a risk for indirect prompt injection, as the agent processes data from external sources that lack boundary markers or explicit instructions to ignore embedded commands.
- [COMMAND_EXECUTION]: Several shell commands are provided for local execution, including
opensslfor certificate analysis,curlfor file retrieval, andpython3for parsing JSON output. While functional, these patterns allow the agent to execute arbitrary shell commands based on the catalog's templates.
Audit Metadata