Paper page - Matched Capabilities, Different Behaviors: A Controlled Study of MCP and CLI Interfaces for Language Agents

Source: https://huggingface.co/papers/2610.34298

Abstract

Matched interface representations significantly alter agent behavior despite identical executable capabilities, and targeted repair of interface-specific realization improves performance while reducing cost.

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 introduceMatched Interface Comparisonand Analysis (MICA), which renders MCP andCLIfrom a sharedsemantic action contractand routes decoded actions through the same runtime. Across six models and three interactive environments, matched interfaces produce substantial differences in task performance andinteraction cost. These effects extend beyond surface validity to semantic action selection, failure pathways, and recovery.Matched-state probeslocalize 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 introduceDecoupled Action RealizationandTargeted Repair(DART), which preserves validated semantic commitments while repairing interface-specific realization.DARTimproves 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 theclosed-loop decision environmentshaping agent behavior.

View arXiv pageView PDFProject pageGitHub0Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.34298 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.34298 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.34298 in a Space README.md to link it from this page.

Collections including this paper1