ctf-web
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill is a technical guide for security research and learning, focusing on CTF challenges. It does not contain malicious instructions, persistence mechanisms, or unauthorized file access.
- [INDIRECT_PROMPT_INJECTION]: The skill involves ingesting untrusted data from external sources, such as analyzing HTML comments, JavaScript files, and server responses from a provided URL. While this represents a data ingestion surface, the instructions guide the agent to perform security analysis rather than executing commands contained within that data.
- [COMMAND_EXECUTION]: The instructions reference the use of common security tools such as
curl,ffuf, andgit-dumperto interact with external target environments. These tools are standard for the pentesting and learning modes specified in the metadata and are used for their intended purpose in a controlled security context.
Audit Metadata