AgentSpawn: Adaptive Multi-Agent Collaboration Through Dynamic Spawning for Long-Horizon Code Generation
Igor Costa · 2026
Read AgentSpawn: Adaptive Multi-Agent Collaboration Through Dynamic Spawning for Long-Horizon Code Generation by Igor Costa (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-07072.txt.
Description
Igor Costa offers AgentSpawn, a design for code generation over long horizons that reshapes its agent team as work unfolds. Long coding jobs need sustained context and shifting expertise across domains. Current multi-agent systems run fixed workflows that cannot adjust when runtime analysis reveals surprise complexity. AgentSpawn enables live teamwork through three mechanisms: memory moves over on its own when a new agent spawns, spawning rules fire on runtime complexity signals, and coherence protocols guard concurrent edits.
The design closes five gaps in prior work. Those gaps are memory continuity and skill inheritance. They also include task resumption and runtime spawning. The fifth is concurrent coherence. In tests it beat fixed baselines by 34 percent on completion rates, on benchmarks including SWE-bench, while trimming memory overhead by 42 percent through selective slicing. The work gives a more flexible template for agent teams facing long, evolving coding jobs.
Catalog record
| CREATOR | Igor Costa |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-07072