The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were detected. The skill is primarily instructional and provides structural context for personal knowledge management.
[COMMAND_EXECUTION]: The skill documents the use of a local obsidian CLI binary and an obsidian-utils script to perform vault operations such as reading, creating, and searching notes. These are standard operations for a file-management skill.
[DATA_EXPOSURE]: The skill hardcodes a local vault path (/Users/kriscard/obsidian-vault-kriscard). This identifies a specific local directory for vault operations but does not involve sensitive system files or credentials.
[INDIRECT_PROMPT_INJECTION]: As the skill reads content from a personal vault, it has an attack surface for indirect prompt injection if external content is saved into notes. The skill mitigates this by instructing the agent to 'Ask, Don't Assume' and requiring explicit user confirmation before writing synthesis or judgments.

vault