The Agent Skills Directory

[COMMAND_EXECUTION]: The skill constructs bash commands by directly interpolating user-provided query strings into templates such as qmd query "<query>". This pattern creates a risk of command injection if the user input contains shell metacharacters (e.g., ;, &, |) that are not properly escaped by the agent before execution.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes and presents untrusted data from an external source (the Obsidian vault).
Ingestion points: Search excerpts and full note content retrieved via qmd query and qmd get commands (SKILL.md).
Boundary markers: Absent. There are no instructions or delimiters defined to help the agent distinguish between note content and its own instructions.
Capability inventory: The skill has the capability to execute various bash commands within the qmd utility suite (SKILL.md).
Sanitization: Absent. The instructions do not specify any validation or sanitization of the content retrieved from the vault before it is processed by the agent.

search-vault