The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes scripts that execute shell commands to modify the host system state by installing packages and updating repositories across multiple platforms. In some cases, these commands are executed with elevated privileges.
Evidence in scripts/install.ps1: Usage of sudo apt-get update and sudo apt-get install -y nm to perform administrative tasks.
Evidence in scripts/install.sh: Platform-specific package management commands including apt-get, dnf, pacman, and apk.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to modify file permissions on the local system.
Evidence in references/troubleshooting.md: Command chmod +r input_file for permission adjustment.
[EXTERNAL_DOWNLOADS]: The installation scripts automate the fetching of software from external system package repositories.
[PROMPT_INJECTION]: The skill is designed to process and analyze binary files from external sources, which constitutes an indirect prompt injection surface if the data is maliciously crafted.
Ingestion points: Input files passed to the nm utility as described in SKILL.md and references/cheatsheet.md.
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are provided in the command patterns.
Capability inventory: The skill has access to shell execution, package management tools, and file permission modification.
Sanitization: No sanitization or validation of the input binary content is performed before processing.

nm