Do LLM Agents Negotiate Rationally? A Mechanism-Design Framework for Verifiable Multi-Agent Interaction over A2A/MCP
Summary
This paper presents a mechanism-design framework for verifying negotiation and allocation tasks in LLM agents using A2A/MCP protocols. It evaluates rational behavior across models, finding that mechanism-level incentive compatibility does not automatically transfer to LLM agents.
View Cached Full Text
Cached at: 08/18/26, 09:52 AM
# Do LLM Agents Negotiate Rationally? A Mechanism-Design Framework for Verifiable Multi-Agent Interaction over A2A/MCP Source: [https://arxiv.org/abs/2608.14613](https://arxiv.org/abs/2608.14613) [View PDF](https://arxiv.org/pdf/2608.14613) > Abstract:Modern LLM\-agent frameworks increasingly interoperate through standards such as Anthropic's Model Context Protocol \(MCP\) for agent\-to\-tool access and Google's Agent2Agent \(A2A\) protocol for agent delegation and negotiation\. However, these protocols specify transport and discovery rather than strategic correctness and do not guarantee efficient, individually rational, or strategy\-proof outcomes\. We introduce a framework that \(i\) encodes classical negotiation mechanisms, including alternating\-offers bargaining and Vickrey\-Clarke\-Groves\-style auctions, as constraints over A2A message schemas; \(ii\) provides a lightweight runtime verification and repair layer that checks messages against protocol invariants; and \(iii\) offers a benchmark of negotiation and allocation tasks with known optimal solutions for measuring deviations from game\-theoretic predictions\. We evaluate multiple LLM backbones using unstructured dialogue, structured protocols, and structured protocols with verification\. Across negotiation trials \(N=30 per condition\), verification reduces outcome variance, while structured protocols achieve 100 percent success for both models\. After correcting parser artifacts, audited unstructured baselines achieve approximately 97 percent and 93\.3 percent success\. In auction experiments \(N=30 per model\), both models achieve 100 percent efficient allocation but differ sharply in truthful bidding: one bids its exact valuation in every trial, whereas the other does so in only 3\.3 percent of trials\. Thus, mechanism\-level incentive compatibility does not automatically transfer to LLM\-agent behavior\. A three\-party fair\-allocation task produced only 4\.2 percent usable outcomes; we report this negative result with a diagnosis\. This work bridges classical multi\-agent systems theory and modern LLM\-agent infrastructure and defines verifiable interaction at the A2A protocol layer\. ## Submission history From: Wael Albayaydh \[[view email](https://arxiv.org/show-email/90261d37/2608.14613)\] **\[v1\]**Fri, 10 Jul 2026 07:32:19 UTC \(529 KB\)
Similar Articles
When LLM Agents Negotiate: Private Information and Dynamic Bargaining in Supply Chains
This paper studies how LLM agents negotiate in a dynamic supply chain bargaining problem, benchmarking nine models from OpenAI, Google, and Alibaba against a Bayesian equilibrium and finding that capability, provider identity, and prompt design shape surplus creation and division.
LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability
This paper formalizes deliberative collaboration for LLM agents under partial observability, introduces a scalable benchmark across multiple domains, and systematically evaluates representative LLMs, finding that complex tasks remain challenging while deliberation can enable error correction.
Investigating Multi-Agent Deliberation in Law
This paper investigates multi-agent deliberation methods for legal reasoning tasks using LLMs, introducing two novel frameworks inspired by courtroom procedures. The experiments show that multi-agent systems achieve comparable overall performance to monolithic LLMs but produce distinct answers and can solve cases that baselines fail, highlighting the potential of multi-agent approaches for legal AI.
Even More Deception: Objective Misalignment in Mixed-Motive LLM Multi-Agent Systems
This paper proposes a framework to evaluate objective misalignment in LLM multi-agent systems using the social deduction game Werewolf, finding that subtle misalignment can profoundly affect collective decision-making.
Dynamic Governance of Multi-LLM Agent Systems for Collaborative Conversational Outcomes
The paper proposes a control-theory-informed governance layer for multi-LLM agent systems, using Contextual Bandit, PID, and POMDP to steer agents toward cooperative outcomes, demonstrating a +32 point lift in simulated financial services interactions.