The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill uses dynamic context injection in the SKILL.md file to verify if the required qmd CLI is installed. It executes qmd status at load time to provide the user with immediate setup feedback.
[UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill instructs the user to install the @tobilu/qmd package from the npm registry. This is an explicit, legitimate dependency required for the skill's primary functionality.
[COMMAND_EXECUTION]: The skill provides numerous examples and tools that involve executing the qmd CLI and curl commands to manage markdown collections and perform local searches.
[INDIRECT_PROMPT_INJECTION]: As a search and retrieval tool for local markdown files, the skill presents a surface for indirect prompt injection if the ingested documents contain malicious instructions that the agent might follow upon reading them.
Ingestion points: Markdown knowledge bases and documentation files added to collections via the qmd collection add command.
Boundary markers: None identified in the provided instructions.
Capability inventory: File system retrieval (get, multi_get), shell command execution (Bash(qmd:*)), and local network operations (curl).
Sanitization: None identified; the skill passes content directly from the search engine to the agent context.

qmd