Runs 7 parallel sub-agents during night shift, each handling different tasks: memory cleanup, budget preparation, TTS research, book recommendations, self-improvement research, AI memory neuroscience, and advisor pattern study. All results committed to git by morning.
Why it matters: Parallel execution multiplies productivity while human sleeps. Different cognitive tasks can run simultaneously without interference.
Real-world example: At 11 PM, parent agent spawns 7 sub-agents. By 6 AM, all complete their tasks and human wakes up to comprehensive briefing.
| Skill | Source | Purpose |
|---|---|---|
sessions_spawn |
Built-in | Create sub-agents |
git |
Built-in | Commit results |
filesystem |
Built-in | Partition memory |
const subAgents = [
{ name: "memory-cleanup", task: "consolidate daily logs" },
{ name: "budget-prep", task: "analyze spending patterns" },
{ name: "tts-research", task: "evaluate new TTS models" },
{ name: "books", task: "find relevant reading" },
{ name: "self-improve", task: "research agent optimization" },
{ name: "neuroscience", task: "study AI memory papers" },
{ name: "advisor", task: "document advisor patterns" }
];## 7-Sub-Agent Night Parallel
At 23:00:
1. Spawn 7 isolated sub-agents with specific tasks
2. Each gets 1-hour timeout
3. Monitor progress every 10 minutes
4. Collect results as each completes
5. Commit all outputs to git
6. Generate consolidated morning briefing
Partitioning rules:
- Each sub-agent has isolated memory/
- No shared write access
- Read-only access to common data- All 7 agents complete within 6 hours
- Results committed to version control
- Morning briefing generated automatically
Example: Clawd42 (Moltbook) - "7 parallel sub-agents"