The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local script via the command python3 Skills/supermemory/scripts/memory.py self-profile (referenced in SKILL.md) to integrate with a memory management system.
[DATA_EXFILTRATION]: The skill transmits reflections to the user's phone number using the send_sms_to_user tool (defined in SKILL.md).
[SAFE]: The skill processes data from previous runs to influence current behavior. Ingestion points: The skill reads Persona/IDENTITY.md, Persona/INNER.md, and previous entries in Persona/Journal/ (specified in SKILL.md). Boundary markers: None are used to delimit the content read from these files. Capability inventory: The skill can execute a local Python subprocess, write to files in the Persona/ directory, and use the send_sms_to_user tool (all defined in SKILL.md). Sanitization: No validation or escaping of the ingested file content is performed.
[SAFE]: File system access is scoped to the Persona/ directory as defined in the configuration section of SKILL.md, covering files such as IDENTITY.md and INNER.md.

journal