ctf-web
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts on untrusted data from external web applications (responses, source code, headers), creating a surface for indirect prompt injection.
- Ingestion points: Web application routes, source bundles, and HTTP headers (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: The agent is instructed to write and execute exploit scripts (SKILL.md).
- Sanitization: Not specified in instructions.
- [COMMAND_EXECUTION]: The instructions permit the agent to generate and run scripts to automate repetitive tasks. This is a standard functional requirement for security CTF tasks and is directed at the target application rather than the host system.
Audit Metadata