The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a custom obsidian command-line tool to perform operations on the host system, including modifying files and controlling a running application instance.- [REMOTE_CODE_EXECUTION]: The obsidian eval command allows for the execution of arbitrary JavaScript within the Obsidian application context. This provides a mechanism for running unvalidated code that can access the internal app object and private APIs.- [DATA_EXFILTRATION]: The skill grants the agent the ability to read any content within the user's Obsidian vaults via obsidian read and obsidian search. This creates a high risk for the unauthorized exposure of sensitive personal information, private notes, or credentials stored within the vault.- [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by reading and processing data from external files (Obsidian notes).
Ingestion points: Vault content is ingested through commands like obsidian read, obsidian daily:read, and obsidian search.
Boundary markers: None observed; the skill does not instruct the agent to delimit or ignore instructions found within note content.
Capability inventory: The agent has capabilities for shell command execution and arbitrary JavaScript execution.
Sanitization: There is no evidence of sanitization or filtering of note content before it is processed by the agent.

obsidian-cli