The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl extensively to perform network requests to the target and external services for testing purposes. These are standard operations for a security testing skill.
[EXTERNAL_DOWNLOADS]: The skill attempts to download and run the interactsh-client binary if it is available in the environment to listen for out-of-band callbacks. It also makes requests to interactsh.com, which is a well-known service for security testing.
[DATA_EXPOSURE]: The skill includes patterns for accessing cloud metadata endpoints (AWS, GCP, Azure) and internal services (Redis, Docker, K8s). While these are sensitive paths, they are accessed on the target server provided by the user for vulnerability assessment, not the user's local system.
[PROMPT_INJECTION]: No prompt injection or behavior override patterns were detected in the instructions.
[OBFUSCATION]: No obfuscated commands, Base64 strings, or hidden URLs were found.

web-vuln-ssrf