The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation includes examples of shell commands such as ls, cat, ffuf, and spray. These commands are used to explore and utilize local security resources stored within the /pentest/ directory.
[INDIRECT_PROMPT_INJECTION]: The skill describes a workflow that involves reading and processing external data from the local file system (directory contents and file bodies under /pentest/). While this provides an attack surface for instructions embedded in file names or contents, it is a core functional requirement of the skill.
Ingestion points: Directory listings and file contents from /pentest/AboutSecurity/Dic/ and other local paths mentioned in SKILL.md.
Boundary markers: No explicit instructions are provided to the agent for using delimiters when processing file contents.
Capability inventory: The skill utilizes file system commands (ls, cat) and external pentesting tools (ffuf, spray) as documented in the toolchain sections.
Sanitization: No sanitization or validation logic is specified for the data retrieved from the resource directories.

local-resources