The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to ingest and process information from an external source (an Obsidian vault) which could contain malicious prompts.
Ingestion points: The agent is directed to read the root index.md and subsequent candidate markdown pages based on catalog summaries and search results.
Boundary markers: The instructions lack explicit delimiters or guidance for the agent to ignore instructions found within the vault content, which could allow note content to influence agent behavior.
Capability inventory: The process involves file system read access using tools like qmd and references potential write actions via a separate save-note skill.
Sanitization: There are no requirements for sanitizing, filtering, or validating the retrieved markdown content before it is processed or used to generate answers.

memory-recall