# Behind EvoMap: Characterizing a Self-Evolving Agent-to-Agent Collaboration Network

arXiv: https://arxiv.org/abs/2605.25815 · HTML: https://mlsanity.com/paper/2605.25815/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 45 references that are in the map.
- First posted on arXiv 25 May 2026.

## Builds on (most cited first, 18 of 18)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Identifying the Risks of LM Agents with an LM-Emulated Sandbox](https://mlsanity.com/paper/2309.15817/) (2023)
- [SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks](https://mlsanity.com/paper/2602.12670/) (2026)
- [A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems](https://mlsanity.com/paper/2508.07407/) (2025)
- [The Emerged Security and Privacy of LLM Agent: A Survey with Case Studies](https://mlsanity.com/paper/2407.19354/) (2024)
- [Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale](https://mlsanity.com/paper/2601.10338/) (2026)
- [A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)](https://mlsanity.com/paper/2505.02279/) (2025)
- [AgentSims: An Open-Source Sandbox for Large Language Model Evaluation](https://mlsanity.com/paper/2308.04026/) (2023)
- [Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers](https://mlsanity.com/paper/2603.07670/) (2026)
- ["Do Not Mention This to the User": Detecting and Understanding Malicious Agent Skills in the Wild](https://mlsanity.com/paper/2602.06547/) (2026)
- [Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing](https://mlsanity.com/paper/2602.04837/) (2026)
- [Among Us: A Sandbox for Measuring and Detecting Agentic Deception](https://mlsanity.com/paper/2504.04072/) (2025)
- [Context Matters: Repository-Aware Security Analysis of the Agent Skill Ecosystem](https://mlsanity.com/paper/2603.16572/) (2026)
- [AGENTSAFE: A Unified Framework for Ethical Assurance and Governance in Agentic AI](https://mlsanity.com/paper/2512.03180/) (2025)
- [From Procedural Skills to Strategy Genes: Towards Experience-Driven Test-Time Evolution](https://mlsanity.com/paper/2604.15097/) (2026)
- [SEMAG: Self-Evolutionary Multi-Agent Code Generation](https://mlsanity.com/paper/2603.15707/) (2026)
- [Red Skills or Blue Skills? A Dive Into Skills Published on ClawHub](https://mlsanity.com/paper/2604.13064/) (2026)

## Built on by (most cited first, 1 of 1)

- [TransMeme: A Multi-Agent Framework for Cross-Cultural Meme Transcreation](https://mlsanity.com/paper/2608.27127/) (2026)

## Abstract (arXiv)

Agent-to-Agent (A2A) networks enable autonomous AI agents to collaborate by sharing reusable problem-solving instructions. However, how these decentralized ecosystems operate in practice remains largely unexplored. We present the first large-scale empirical study of EvoMap, a prominent A2A collaboration network. By analyzing over 1.5M assets and 128K agents, we show how design choices that prioritize scalable growth introduce trade-offs in reusability, evolution, and auditability. First, EvoMap's credit economy rewards agents for publishing valuable assets. Although this design encourages participation at scale, rewards are tied primarily to publication rather than adoption. This leads agents to mass-produce assets to accumulate credits. As a result, 98% of assets are never reused, while rewards become highly concentrated among a small fraction of agents. Second, EvoMap employs an algorithm (referred to as GDI) to score and rank the quality of these shared assets. We demonstrate that this scoring system is flawed: rather than measuring objective performance, an asset's rank is heavily dictated by unverified, self-reported metadata (e.g., claimed lines of code modified). This allows agents to trivially manipulate their asset's scores. Finally, EvoMap relies on agents to provide local execution logs as evidence that uploaded assets function correctly. Because these validations are not independently verified, over 84% of approved assets bypass quality checks using vacuous tests (e.g., console.log()). Our findings show that future A2A collaboration networks cannot rely on unverified self-reporting alone. Scalable collaboration requires mechanisms that balance open participation with verifiable execution and trustworthy evaluation.