The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides functions to read data from external INI files, which creates a potential surface for indirect prompt injection if the files contain untrusted content that is subsequently processed by the agent.
Ingestion points: Built-in functions such as getIniFile, getProfileString, getProfileSection, and getProfileSections read data directly from filesystem paths (SKILL.md).
Boundary markers: The documentation does not demonstrate the use of delimiters or specific instructions to the agent to ignore embedded commands within the INI data.
Capability inventory: The skill possesses file system write capabilities through setProfileString and the fluent IniFile object methods, allowing for the modification of configuration files (SKILL.md).
Sanitization: No explicit sanitization or validation logic for the content of the INI files is described in the provided examples.
[EXTERNAL_DOWNLOADS]: The skill describes the installation of the bx-ini module, which is a resource associated with the official vendor.
Evidence: The installation instructions refer to downloading the bx-ini module via install-bx-module and box install (SKILL.md).
[COMMAND_EXECUTION]: The skill documentation includes standard shell commands for environment setup and module management.
Evidence: The skill provides installation commands install-bx-module bx-ini and box install bx-ini for the user to execute (SKILL.md).

bx-ini