What problem does it solve?
Running recurring maintenance tasks like security audits, performance optimization, and memory consolidation manually is repetitive and easy to forget. This Skill automates periodic background workers inside a Claude Code session using native /loop scheduling with cache-aware delays.
Core Features & Use Cases
- Periodic Worker Dispatch: Trigger named workers (audit, optimize, consolidate, predict, map, testgaps, document, benchmark) via the Ruflo MCP worker-dispatch tool.
- Cache-Aware Scheduling: Computes delays with the formula min(270, cache_ttl * 0.9) to keep the prompt cache warm between iterations.
- Self-Continuing Loops: Uses ScheduleWakeup to queue the next iteration automatically, following [LOOP_SUGGESTION] and [CRON_SUGGESTION] hints in MCP responses.
- Use Case: Start the audit worker to run a security analysis every 270 seconds during a long coding session, with each iteration rescheduling itself until the session ends.
Quick Start
Ask the AI to run the audit background worker on a loop so it performs security analysis every 270 seconds during this session.