The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like find and grep to interact with a hardcoded local directory (/mnt/d/Obsidian Vault/AI Research/).
Evidence: Shell workflows in SKILL.md demonstrate direct interpolation of search keywords and note titles into command lines.
Risk: If input is not properly sanitized by the platform, there is a potential for command injection via shell metacharacters.
[PROMPT_INJECTION]: The skill implements an indirect prompt injection surface by ingesting the content of markdown notes into the agent's context during search and retrieval operations.
Ingestion points: Note content is read via grep -rl and similar commands in SKILL.md.
Boundary markers: None identified; external note content is processed without explicit delimiters or warnings to ignore embedded instructions.
Capability inventory: The skill has the capability to execute shell commands (find, grep) as described in SKILL.md.
Sanitization: There is no evidence of sanitization, validation, or escaping of note content before it is processed by the agent.

obsidian-vault