The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the obsidian CLI tool and standard shell commands (like open and sleep) to interact with the Obsidian application and manage vault files. These operations are strictly limited to the intended productivity workflow and do not involve arbitrary system modification.\n- [PROMPT_INJECTION]: The skill's 'Clipping Save' and 'Knowledge Organizer' features process external web content, creating a surface for indirect prompt injection. (1) Ingestion points: Data from the search-and-fetch skill and files in the 31_WebClips/ folder. (2) Boundary markers: The agent is instructed to respect privacy tags (#Private, #Key), but no specific escape delimiters are used for web content. (3) Capability inventory: Extensive file interaction capabilities via the obsidian CLI. (4) Sanitization: The skill mandates a summarization and knowledge extraction process, which naturally filters content through the agent's reasoning before it is saved.\n- [SAFE]: The skill implements clear privacy boundaries and a 'read-before-write' protocol (Section Editing Protocol), ensuring that the agent does not modify sensitive information or inadvertently overwrite important user data.

obsidian-knowledge