The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied flashcard data, which constitutes an untrusted input surface.
Ingestion points: The skill reads flashcard prompts, answers, and hierarchical structures (rem) from user input or connected databases.
Boundary markers: The instructions include a logical constraint requiring the agent to only write changes when the "destination and requested change are clear," though no technical delimiters or specific safety warnings against embedded instructions are provided.
Capability inventory: The skill possesses the capability to report findings and potentially modify external data stores via connected RemNote tools.
Sanitization: There is no evidence of sanitization or filtering for potentially malicious instructions embedded within the flashcard content being reviewed.
[EXTERNAL_DOWNLOADS]: The skill provides links to official help articles on the help.remnote.com domain. These are informational references to a well-known service intended for user education and do not involve the automated execution of remote scripts or the download of executable binaries.

remnote-flashcard-reviewer