Beyond Prompt-Based Planning: MCP-Native Graph Planning-based Biomedical Agent System

arXiv 2606.04494 ↗First posted 3 Jun 2026cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Biomedical agents promise to automate complex biological workflows, yet current systems face two fundamental bottlenecks: bioinformatics tools are highly heterogeneous in interfaces and execution environments, while agent planning still relies on flat prompt-retrieved tool descriptions. As biomedical software ecosystems grow, this coupling between tool coverage and context size leads to tool confusion, unstable planning, and inefficient execution. We introduce BioManus, an MCP-native biomedical agent built on graph-scaffolded planning over structured biological capabilities. BioManus first introduces the BioinfoMCP Compiler, which converts heterogeneous bioinformatics software into standardized MCP servers, yielding a large executable MCP ecosystem. It then organizes this ecosystem as a typed heterogeneous MCP graph over tools, operations, datatypes, and workflow stages. At inference time, BioManus retrieves compact task-specific subgraphs, synthesizes operation-level workflow scaffolds. This design decouples planning complexity from raw tool inventory size, achieving a context compression ratio of Theta(N / (h * m_bar)) under high-recall retrieval, where N is the total tool count, h is the workflow horizon, and m_bar (much smaller than N) is the average number of candidate tools per operation. Experiments on BioAgentBench and LAB-Bench show that BioManus improves execution accuracy, workflow validity, and context efficiency over advanced biomedical agent baselines. This work suggests a paradigm shift: scalable biomedical reasoning requires structured executable capability graphs rather than increasingly larger prompt-level tool retrieval.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 17 of its 166 references that are in the map. First posted on arXiv 3 Jun 2026.

Builds on

17 of 166 references.

References · 17

2020Language Models are Few-Shot Learnerscited by 32,199
2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
2023Gorilla: Large Language Model Connected with Massive APIscited by 1,302
2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligencecited by 877
2024LAB-Bench: Measuring Capabilities of Language Models for Biology Researchcited by 138
2023RestGPT: Connecting Large Language Models with Real-World RESTful APIscited by 129
2023GeneGPT: Augmenting Large Language Models with Domain Tools for Improved Access to Biomedical Informationcited by 115
2015Yara Parser: A Fast and Accurate Dependency Parsercited by 110
2025MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Serverscited by 101
2024CellAgent: An LLM-driven Multi-Agent Framework for Automated Single-cell Data Analysiscited by 38

5 more

2023Prompt Sapper: A LLM-Empowered Production Tool for Building AI Chainscited by 24
2024GeneAgent: Self-verification Language Agent for Gene Set Knowledge Discovery using Domain Databasescited by 8
2026BioAgent Bench: An AI Agent Evaluation Suite for Bioinformaticscited by 8
2026ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandboxcited by 3
2026Biomedical systems biology workflow orchestration and execution with PoSyMedcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way