The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function involves executing shell commands to inspect files and manage system packages. Specifically, scripts/install.ps1 uses sudo to escalate privileges for updating and installing packages on Linux systems.
[COMMAND_EXECUTION]: Documentation in references/troubleshooting.md instructs the agent to use chmod +r to modify file permissions, which involves direct system manipulation.
[PROMPT_INJECTION]: The skill facilitates the ingestion of external, potentially untrusted binary data through hexdump. This creates a surface for indirect prompt injection where malicious instructions embedded in processed files could influence agent behavior.
Ingestion points: Files provided by users are processed via hexdump in SKILL.md and references/cheatsheet.md.
Boundary markers: The instructions do not define clear delimiters or warnings to ignore instructions found within the data being inspected.
Capability inventory: The skill has access to shell execution for tool installation and file inspection.
Sanitization: There is no evidence of sanitization or filtering of the file content before it is displayed to the agent.

hexdump