The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use an external obsidian CLI tool to perform file operations and application management tasks. \n- [REMOTE_CODE_EXECUTION]: The obsidian eval command allows for arbitrary JavaScript execution within the Obsidian application context. This presents a high-risk capability that could be abused to perform unauthorized actions or access sensitive data. \n- [PROMPT_INJECTION]: The skill facilitates data ingestion that is vulnerable to indirect prompt injection: \n
Ingestion points: Operations such as obsidian read, obsidian search, and obsidian daily:read allow untrusted content from vault notes to enter the agent's context. \n
Boundary markers: There are no instructions or markers provided to help the agent distinguish between its own system instructions and potentially malicious instructions contained within notes. \n
Capability inventory: The agent has extensive capabilities including file system modification, application control through eval, and visual inspection via obsidian dev:screenshot. \n
Sanitization: No evidence of sanitization or content validation is present to mitigate the risks of processing third-party data.

obsidian-cli