The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the obsidian command-line tool to interact with a running Obsidian instance. This includes commands for reading, creating, and appending to files (obsidian read, obsidian create, obsidian append), as well as administrative tasks like reloading plugins.
[COMMAND_EXECUTION]: The skill provides an obsidian eval command which allows the execution of arbitrary JavaScript code directly within the Obsidian application's environment. This provides broad access to the application's internal API and data.
[PROMPT_INJECTION]: The skill acts as an interface for reading untrusted content from an Obsidian vault (via read, search, and dev:console commands). This content is then processed by the agent, creating a surface for indirect prompt injection if the vault data contains malicious instructions intended to manipulate the agent's behavior.
[SAFE]: The skill references official documentation from help.obsidian.md, which is a well-known and trusted source for this application.

obsidian-cli