The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill provides an obsidian eval command which allows the execution of arbitrary JavaScript code strings within the context of the running Obsidian application. This feature is intended for plugin development and advanced vault interrogation (e.g., app.vault.getFiles().length), but it grants the agent the ability to run unvalidated code that could interact with the underlying system or vault data in unintended ways.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a clear indirect prompt injection attack surface by reading untrusted data from the user's vault.
Ingestion points: Vault content is ingested via obsidian read, obsidian search, obsidian daily:read, and obsidian backlinks.
Boundary markers: The instructions do not define delimiters or specific warnings for the agent to ignore instructions embedded within the vault files.
Capability inventory: The skill has high-impact capabilities including writing files (create, append, property:set) and executing arbitrary JavaScript code (eval).
Sanitization: There are no mentioned mechanisms for sanitizing or validating the content retrieved from notes before the agent processes it.
[COMMAND_EXECUTION]: The skill facilitates the execution of various CLI commands that modify the local file system and application state. While these are the primary functions of the skill, the lack of constraints on the parameters passed to these commands could lead to unintended file modifications or deletions if the agent is influenced by malicious input.

obsidian-cli