Subagent Fan-out
SkillAI & modelssubagent-fanout lets your AI split a goal into independent pieces and work through them in parallel, each piece in its own fresh context. Instead of piling every detail into one long conversation, separate sub-jobs run side by side and stay focused. This fits tasks that branch into many parts that do not depend on each other.
Available today. Use it from your connected AI after setup.
No other account needed.
After adding it, describe a goal that breaks into independent pieces and ask your AI to run the pieces in parallel. The skill takes care of fanning the sub-jobs out from there.
Then ask your AI: use the Subagent Fan-out skill
What your AI can do with it
- Split a goal into independent pieces and run them in parallel
- Give each sub-job its own fresh context instead of one crowded conversation
- Work through many sub-jobs at the same time
- Keep the main conversation from filling up with every detail
- Handle tasks that branch into several unrelated parts
What this skill tells your AI
The instructions your AI receives, as published by archive228/loopkit in skills/subagent-fanout/SKILL.md and read by ahel’s review.
One context loaded with ten jobs' worth of material is the exact shape that triggers context rot. Ten small contexts don't.
- Spawn one subagent per independent unit (one file, one source, one check). Each gets a fresh context window.
- An orchestrator synthesizes their results — it never does the per-unit work itself.
- Give each worker a tight role and only the input it needs.
- Use ONLY when the pieces are genuinely independent. Sequential dependencies stay in one chain. Fan-out for breadth (research, multi-file edits, multi-source verification). Keep it serial when step N needs step N-1's output.
Signals
- GitHub stars
- 755
- Forks
- 125
- Last commit
- Jul 2026
Advanced
- Catalog kind
- skill
- Gateway key
subagent-fanout- Source
- github.com/archive228/loopkit