arXiv cs.AI· Zhenning Shi, Liang Pang, Yongqi Li·· 12 小时前AI 评分45
MCP 与 CLI 接口对比研究：能力匹配下语言智能体行为差异及 DART 修复方法
Matched Capabilities, Different Behaviors: A Controlled Study of MCP and CLI Interfaces for Language Agents
AI 导读
研究提出 MICA 框架，在能力匹配条件下对比 MCP 与 CLI 接口对语言智能体的影响，发现六款模型在三个交互环境中任务表现与交互成本差异显著，2,004 条配对轨迹中 94.3% 出现结构化分歧。基于此提出 DART 方法，在保留已验证语义承诺的同时修复接口特定实现，六个评测设置基准分数最高提升 9.0 分，三个环境中 MCP token 成本每任务最多降低 27.1k。
正文
Abstract:Language agents increasingly act through external tools and interactive environments, making the interface between the model and its executable capabilities part of the decision process. Model Context Protocol (MCP) and command-line interfaces (CLI) can expose the same semantic operations through markedly different model-facing representations. We study whether these representational differences change agent behavior when executable capabilities are matched. We introduce Matched Interface Comparison and Analysis (MICA), which renders MCP and CLI from a shared semantic action contract and routes decoded actions through the same runtime. Across six models and three interactive environments, matched interfaces produce substantial differences in task performance and interaction cost. These effects extend beyond surface validity to semantic action selection, failure pathways, and recovery. Matched-state probes localize sensitivity to capability presentation, realization, and feedback. Paired natural trajectories reveal structured divergence in 94.3% of 2,004 episodes and show that early differences propagate through subsequent interaction. This analysis identifies interface realization as a direct intervention boundary. We therefore introduce Decoupled Action Realization and Targeted Repair (DART), which preserves validated semantic commitments while repairing interface-specific realization. DART improves benchmark score in all six evaluated settings by up to 9.0 points and reduces MCP token cost in all three environments by up to 27.1k tokens per task. Together, these results establish interface representation as part of the closed-loop decision environment shaping agent behavior.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.34298 [cs.AI] |
| (or arXiv:2610.34298v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34298 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhenning Shi [view email]
[v1]
Sun, 25 Oct 2026 13:54:07 UTC (551 KB)
来源：arXiv cs.AI · arxiv.org