The Agent Skills Directory

[SAFE]: No malicious code or unauthorized data access patterns were detected. The skill performs natural language parsing and schedules tasks using standard platform-provided tools.
[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface as it captures user-provided text for future delivery. This is inherent to the skill's purpose and is mitigated by platform-level isolation.
Ingestion points: User input is parsed in SKILL.md to extract the message content.
Boundary markers: The scheduled payload in SKILL.md and references/TEMPLATES.md uses a "REMINDER: " prefix for the message.
Capability inventory: The skill invokes the cron.add tool to schedule agentTurn payloads.
Sanitization: No explicit sanitization or escaping of the user's message is performed before it is scheduled.

remindme