# AgentSpawn: Adaptive Multi-Agent Collaboration Through Dynamic Spawning for Long-Horizon Code Generation

arXiv: https://arxiv.org/abs/2602.07072 · HTML: https://mlsanity.com/paper/2602.07072/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Feb 2026.

## Built on by (most cited first, 8 of 8)

- [Natural-Language Agent Harnesses](https://mlsanity.com/paper/2603.25723/) (2026)
- [Scaling Discovery through Test-Time Communication](https://mlsanity.com/paper/2609.21032/) (2026)
- [Reinforcement Learning for LLM-based Multi-Agent Systems through Orchestration Traces](https://mlsanity.com/paper/2605.02801/) (2026)
- [SkillMAS: Skill Co-Evolution with LLM-based Multi-Agent System](https://mlsanity.com/paper/2605.09341/) (2026)
- [Spawn Freely, Act Sparingly: Progressive Risk Vesting for Recursive LLM-Agent Trees](https://mlsanity.com/paper/2609.01035/) (2026)
- [PrimeAgentOrchestrator: Memory-Primed Agent Spawning for Personal AI Infrastructure](https://mlsanity.com/paper/2608.20342/) (2026)
- [AsynCodeBench: Benchmarking Collaboration of Asynchronous Multi-Agent Systems in Software Engineering](https://mlsanity.com/paper/2609.32662/) (2026)
- [Peer Influence across Heterogeneous AI Models](https://mlsanity.com/paper/2610.03095/) (2026)

## Abstract (arXiv)

Long-horizon code generation requires sustained context and adaptive expertise across domains. Current multi-agent systems use static workflows that cannot adapt when runtime analysis reveals unanticipated complexity. We propose AgentSpawn, an architecture enabling dynamic agent collaboration through: (1) automatic memory transfer during spawning, (2) adaptive spawning policies triggered by runtime complexity metrics, and (3) coherence protocols for concurrent modifications. AgentSpawn addresses five critical gaps in existing research around memory continuity, skill inheritance, task resumption, runtime spawning, and concurrent coherence. Experimental validation demonstrates AgentSpawn achieves 34% higher completion rates than static baselines on benchmarks like SWE-bench while reducing memory overhead by 42% through selective slicing.