The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill outlines a process for fingerprinting Web Application Firewalls by analyzing responses from external HTTP services, which represents an attack surface where a target could attempt to influence the agent via crafted headers or response bodies.
Ingestion points: HTTP responses from external target servers analyzed by the tool (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided.
Capability inventory: Network communication (HTTP probes) and data parsing.
Sanitization: Not explicitly handled in the documentation; safety relies on the underlying tool's handling of response data.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute the wafw00f command-line utility with various arguments for scanning targets and outputting results. These commands are consistent with the tool's primary purpose and do not exhibit malicious intent.
[SAFE]: The skill content is authored by AeonDave, the recognized author of the wafw00f project. All described functionality is standard for security reconnaissance and does not include obfuscation, persistence, or data exfiltration patterns.

wafw00f