The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill configuration specifies the installation of the qmd CLI tool directly from a GitHub repository (https://github.com/tobi/qmd) via the Node.js package manager.
[COMMAND_EXECUTION]: The skill relies on the execution of the local qmd binary to manage file collections, update indexes, and perform searches.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes content from local files that may contain adversarial instructions.
Ingestion points: Local files indexed using the qmd collection add command and retrieved via qmd search or qmd get.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the retrieved file content.
Capability inventory: The skill executes shell commands (qmd) and interacts with a local embedding service (Ollama).
Sanitization: There is no documented validation or filtering mechanism to prevent the agent from obeying instructions found inside the indexed documents.

qmd