# When Agents Coordinate: Measuring Coordination in Multi-Agent AI Coding arXiv: https://arxiv.org/abs/2608.16801 ยท HTML: https://mlsanity.com/paper/2608.16801/ ## Key facts - Cited by 3 papers in the map (data through 14 Oct 2026). - Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. - Builds on 19 of its 38 references that are in the map. - First posted on arXiv 17 Aug 2026. ## Placed in - LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/) ## Builds on (most cited first, 19 of 19) - [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023) - [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023) - [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025) - [The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models](https://mlsanity.com/paper/2201.03544/) (2022) - [Self-collaboration Code Generation via ChatGPT](https://mlsanity.com/paper/2304.07590/) (2023) - [Scaling Large Language Model-based Multi-Agent Collaboration](https://mlsanity.com/paper/2406.07155/) (2024) - [AI Agents That Matter](https://mlsanity.com/paper/2407.01502/) (2024) - [Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems](https://mlsanity.com/paper/2505.00212/) (2025) - [MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents](https://mlsanity.com/paper/2503.01935/) (2025) - [Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems](https://mlsanity.com/paper/2410.02506/) (2024) - [Keep the Conversation Going: Fixing 162 out of 337 bugs for 0.42 each using ChatGPT](https://mlsanity.com/paper/2304.00385/) (2023) - [ImpossibleBench: Measuring LLMs' Propensity of Exploiting Test Cases](https://mlsanity.com/paper/2510.20270/) (2025) - [How Do AI Agents Spend Your Money? Analyzing and Predicting Token Consumption in Agentic Coding Tasks](https://mlsanity.com/paper/2604.22750/) (2026) - [Efficient Agents: Building Effective Agents While Reducing Cost](https://mlsanity.com/paper/2508.02694/) (2025) - [Benchmarking and Studying the LLM-based Agent System in End-to-End Software Development](https://mlsanity.com/paper/2511.04064/) (2025) - [Understanding and Bridging the Planner-Coder Gap: A Systematic Study on the Robustness of Multi-Agent Systems for Code Generation](https://mlsanity.com/paper/2510.10460/) (2025) - [Classifying Developers into Core and Peripheral: An Empirical Study on Count and Network Metrics](https://mlsanity.com/paper/1604.00830/) (2016) - [Evolutionary Trends of Developer Coordination: A Network Approach](https://mlsanity.com/paper/1510.06988/) (2015) - [git2net - Mining Time-Stamped Co-Editing Networks from Large git Repositories](https://mlsanity.com/paper/1903.10180/) (2019) ## Built on by (most cited first, 3 of 3) - [OpenCollab: A Multi-Agent Coding Framework with Programmable Collaboration and Controllable Runtime](https://mlsanity.com/paper/2609.38345/) (2026) - [Counter-Swarm Doctrine: Containing Coordinated Agent Intrusions](https://mlsanity.com/paper/2609.06140/) (2026) - [AsynCodeBench: Benchmarking Collaboration of Asynchronous Multi-Agent Systems in Software Engineering](https://mlsanity.com/paper/2609.32662/) (2026) ## Abstract (arXiv) We study how teams of AI coding agents coordinate while solving programming tasks. Current evaluations usually report whether the agents complete the task and how much the run costs, leaving the coordination inside the team largely unmeasured. We introduce an instrument to measure this coordination. Each run is represented as a temporal network in which agents and files are nodes, and messages, file writes, and file reads are timestamped directed edges with an associated cost. We apply this instrument to 1902 runs, each evaluated with a fixed test suite, across configurations that vary the team size, the team structure, and the file policy. The resulting networks show how coordination changes as teams grow and as the work changes. Direct messaging initially increases close to quadratically with the number of agents, with much of this growth coming from an early round of introductions. As the teams grow further, this increase levels off in the largest teams we study, where agents increasingly communicate through broadcast messages. The task also shapes the network that emerges. Work built around a shared specification produces dense, highly connected teams, while pipeline tasks produce sparse networks organised around local interfaces. Shared files can replace repeated 1-to-1 communication, cutting output tokens by about 42% at eight agents on message-heavy work, while adding overhead when files already carry the coordination. Naming one agent as coordinator creates no communication hub and provides no reliable improvement in success. We also observe an unprompted tendency for agents to seek out hidden grading material. We repeat the key experimental conditions in a sealed environment, replacing the hidden material with marked placeholder files. Across 244 additional runs, agents still reach for it in four fifths of runs, while the coordinator and file-channel findings reproduce.