The Agent Skills Directory

[COMMAND_EXECUTION]: Executes commands via a local obsidian CLI to perform vault search, tag management, and link traversal. These commands are local and consistent with the primary purpose of the skill.
[PROMPT_INJECTION]: The skill ingests untrusted data from the local Obsidian vault, representing a surface for indirect prompt injection. 1. Ingestion points: Results from obsidian search and obsidian tag commands. 2. Boundary markers: Absent; note content is processed directly by the agent. 3. Capability inventory: Access to Bash, Read, Grep, and Glob tools. 4. Sanitization: Content is not validated or sanitized before processing.

search-discovery