The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the qmd CLI (or an overridden path in QMD_CLI) to update and verify a search index after vault writes.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user's Obsidian vault (index.md) to understand vault structure, which creates a potential surface for indirect prompt injection.
Ingestion points: Reading $OBSIDIAN_VAULT_PATH/index.md (SKILL.md, Step 2).
Boundary markers: None specified for file ingestion.
Capability inventory: File write operations to .base, .md, and log.md files; execution of qmd CLI commands (update, embed, ls, get).
Sanitization: None mentioned for processing external markdown content.

wiki-dashboard