The Agent Skills Directory

[SAFE]: The skill performs legitimate configuration checks using official tools (list_workspaces, list_memory_projects). It confirms the assistant is connected to the vendor's cloud service and requires explicit user approval before writing any data.
[INDIRECT_PROMPT_INJECTION]: The skill ingests a verification code from the user. It mitigates potential injection risks by instructing the agent to validate the code's format (prefix and length) and specifically mandates that the data be stored in metadata fields rather than the note's content body to prevent interpretation as instructions.
[DATA_EXFILTRATION]: The skill's primary function involves saving a verification code to the vendor's cloud service (basicmemory.com). This is the intended purpose of the mission and does not involve unauthorized access to sensitive local files, credentials, or system environment data.

memory-quest