The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the obsidian CLI and the open command to interact with the host system and the Obsidian application to manage notes and application state.
[DYNAMIC_EXECUTION]: The obsidian eval command allows the agent to execute arbitrary JavaScript code within the context of the running Obsidian application, facilitating the execution of dynamic logic at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from vault notes which may contain malicious instructions that could influence agent behavior.
Ingestion points: Note content retrieved using the obsidian read and obsidian search commands in SKILL.md.
Boundary markers: The instructions lack delimiters or explicit directives for the agent to ignore or isolate instructions found within note content.
Capability inventory: JavaScript execution (obsidian eval), file writing/modification (obsidian create, obsidian append), and application state inspection (obsidian dev:console, obsidian dev:screenshot) in SKILL.md.
Sanitization: No evidence of input validation or content sanitization exists for the data retrieved from the vault.

obsidian