The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill documents the obsidian eval command, which permits the execution of arbitrary JavaScript within the running Obsidian application environment. While intended for developer workflows and automation, this capability allows for dynamic code execution at runtime.
[COMMAND_EXECUTION]: The skill utilizes the obsidian command-line interface to interact with vaults, enabling actions such as reading/writing notes, searching vault content, and performing developer tasks like capturing screenshots and inspecting the application's DOM.
[EXTERNAL_DOWNLOADS]: The instructions recommend installing the mcp-obsidian-cli tool using npm or npx to provide an Model Context Protocol (MCP) server interface, facilitating the agent's interaction with the Obsidian vault.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from Obsidian notes, which acts as an untrusted data source that could influence agent behavior.
Ingestion points: Note content is retrieved via obsidian read, obsidian search, and obsidian daily:read commands in the references/obsidian-cli.md file.
Boundary markers: No specific delimiters or "ignore instructions" markers are defined for the ingested content.
Capability inventory: The skill possesses file system modification capabilities (create, append, property:set) and application-level JavaScript execution (eval) as documented in references/obsidian-cli.md.
Sanitization: No specific content sanitization, escaping, or validation logic is implemented for the processed data.

obsidian