The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the find and grep utilities to search for filenames and textual content within the local directory path /mnt/d/Obsidian Vault/AI Research/. These commands are scoped to the vault's location for its primary note-management functions.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests and processes content from user-generated Markdown files. 1. Ingestion points: External note content is read into the agent's context from the local vault via search and retrieval workflows. 2. Boundary markers: No specific markers or delimiters are defined to isolate note content or instruct the agent to ignore embedded instructions. 3. Capability inventory: The skill can read, search, and manage files within the local directory. 4. Sanitization: There are no explicit content validation or filtering steps mentioned for the data retrieved from the vault.
[SAFE]: No obfuscation, data exfiltration, or unauthorized credential access patterns were detected. The skill's behavior is consistent with its stated purpose of managing a personal knowledge base.

obsidian-vault