mlsanityData to 14 Oct 2026

Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions

arXiv 2610.14334 ↗First posted 9 Oct 2026cs.AI cs.CL cs.SEcited by 0

On the map: LLM agents and tool use → Tool learning and function calling

Abstract (arXiv)

In Model Context Protocol (MCP) systems, LLMs select tools from metadata rather than from their implementations. Names and descriptions summarize what a tool does, while the parameter schema specifies the fields a tool accepts and how they are organized. Our analysis of metadata for 367,174 MCP tools reveals substantial schema variation among functionally similar tools. Many of these tools share names and similar descriptions but expose different parameter schemas. This raises a key question: can parameter-schema differences influence tool selection before any call is made? We study this question through controlled metadata interventions across five LLMs and five tasks. Each comparison changes exactly one metadata factor (server name, tool name, description, or parameter schema) while keeping the others fixed. We balance tool positions and use conditional logistic regression to estimate position-adjusted effects. We find that: (1) parameter-schema changes produce the largest and most consistent selection shifts among the tested interventions. Models tend to favor separate top-level parameters over grouped or composite forms. (2) Descriptions that explain parameters also tend to increase selection, while name effects are weaker and less consistent. (3) Targeted comparisons show that schema effects depend on the specific parameter names and on how the request represents the same information. When only one schema satisfies an explicit request requirement, all five models select the compatible tool despite position preferences. These results show that parameter schemas influence not only how tools are called, but also whether they are selected. For MCP developers and evaluators, schema design and tool ordering can therefore affect which implementation is invoked even when functionality is unchanged.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use. Builds on 21 of its 27 references that are in the map. First posted on arXiv 9 Oct 2026.

Builds on

21 of 27 references.

Closest ancestors · 10

Same direction, newest first

  1. 2026Diagnosing Tool-Selection Reasoning in LLM Agents with Canary Tools5 Aug
  2. 2026Model Context Protocol (MCP) Tool Descriptions Are Smelly! Towards Improving AI Agent Efficiency with Augmented MCP Tool Descriptions16 Feb
  3. 2025MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Servers28 Aug
  4. 2025MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Servers20 Aug
  5. 2025Tool Preferences in Agentic LLMs are Unreliable23 May
  6. 2025RAG-MCP: Mitigating Prompt Bloat in LLM Tool Selection via Retrieval-Augmented Generation6 May
  7. 2025On the Robustness of Agentic Function Calling1 Apr
  8. 2024RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning16 Jan
  9. 2024EASYTOOL: Enhancing LLM-based Agents with Concise Tool Instruction11 Jan
  10. 2023MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use4 Oct

Other references · 11

  1. 2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligencecited by 1,040
  2. 2023Large Language Models Are Not Robust Multiple Choice Selectorscited by 486
  3. 2023Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questionscited by 263
  4. 2025Prompt Injection Attack to Tool Selection in LLM Agentscited by 158
  5. 2023Tool Documentation Enables Zero-Shot Tool-Usage with Large Language Modelscited by 68
  6. 2025MPMA: Preference Manipulation Attack Against Model Context Protocolcited by 34
  7. 2024Manipulating Large Language Models to Increase Product Visibilitycited by 28
  8. 2025ToolTweak: An Attack on Tool Selection in LLM-based Agentscited by 26
  9. 2025What Is Your AI Agent Buying? Evaluation, Biases, Model Dependence, & Emerging Implications for Agentic E-Commercecited by 22
  10. 2025BiasBusters: Uncovering and Mitigating Tool Selection Bias in Large Language Modelscited by 11
  11. 2025A First Look at the Security Issues in the Model Context Protocol Ecosystemcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026LLM within MCP Matters: Measuring Inefficient Resource Utilization Driven by LLMs
  2. 2026Understanding How Enterprises Adopt the Model Context Protocol for LLM-Driven Software Engineering
  3. 2026MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers
  4. 2025MCP-RADAR: A Multi-Dimensional Benchmark for Evaluating Tool Use Capabilities in Large Language Models
  5. 2025Help or Hurdle? Rethinking Model Context Protocol-Augmented Large Language Models
  6. 2025ScaleMCP: Dynamic and Auto-Synchronizing Model Context Protocol Tools for LLM Agents
  7. 2025LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?
  8. 2026Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use