Loop Worker

This sets up background workers that run on Claude Code's native loop scheduling, keeping your prompt cache warm with carefully timed 270 second intervals. You dispatch workers through MCP hooks for things like security audits, performance optimization, or test coverage analysis, then schedule the next iteration with ScheduleWakeup. The delay formula is tuned to stay under cache TTL so you're not burning tokens on cold starts. Eight preconfigured workers are included, from predictive preloading to auto-documentation. If you're already using Ruflo and want continuous background tasks without manual triggering, this handles the scheduling mechanics for you.

skills.sh installs: 745
GitHub stars: 81.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add ruvnet/ruflo --skill loop-worker --agent claude-code

Installs into .claude/skills of the current project.

Files

Use /loop for in-session periodic workers. Pick the delay from what you are waiting for:

With the 5-minute prompt-cache TTL, min(270, cache_ttl * 0.9) = 270s keeps the cache warm (ruflo-autopilot ADR-0001 owns that heartbeat).
If the ScheduleWakeup tool description says this session uses the 1-hour TTL, a 270s wake buys nothing; poll at 1200-1800s unless the worker finishes faster.

Dispatch a worker via MCP: mcp__plugin_ruflo-core_ruflo__hooks_worker-dispatch({ trigger: "WORKER_NAME" }). It runs in the background by default (background: true), so check on it with hooks_worker-status instead of waiting inline.

Then schedule the next iteration. ScheduleWakeup needs the loop prompt back so the next firing re-enters the loop: ScheduleWakeup({ delaySeconds: 270, reason: "next WORKER_NAME iteration", prompt: "/loop <the same /loop input>", noop: false })

Stop the loop when the worker reports nothing new twice in a row: ScheduleWakeup({ stop: true }).