Skip to content

An open project initiated by the DeepModeling community focused on constructing intelligent agents for scientific research. It is committed to assembling a reusable and extensible set of Agent-Ready tools, promoting the development and practical deployment of research agents through open collaboration.

License

Notifications You must be signed in to change notification settings

deepmodeling/build-your-agent

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 

Repository files navigation

build-your-agent

An open project initiated by the DeepModeling community focused on constructing intelligent agents for scientific research.

We continuously release and share a collection of agent samples that are oriented towards real scientific research problems, to help developers understand practical agent architecture design and construction methods. These examples aim to span both user-friendly instructions and multiple cutting-edge application scenarios.

Our recent goals include:

  • Define and validate a reference architecture for “scientific research agents”
    (Perception → Planning → Execution → Feedback)
  • Build sample agents for typical scientific scenarios:
    • Domain-Specific Research Assistant: Retrieve and summarize literature based on keywords, and automatically generate review drafts
    • Materials Design Optimization Agent: Structure generation → Simulation calculation → Performance evaluation → Structure screening
      • dpa_calculator An agent to use universal potential DPA-2.4 to automatically relax a materials structure and calcualte its property ( such as phonon).
    • Drug Molecule Screening Agent: Molecule filtering and optimization process based on multiple drug-likeness properties
    • Omics Data Intelligent Analysis Agent: Integrate cross-domain data and tools to automatically complete result analysis
    • More ideas are welcome!
  • Provide agent development guides and tutorial notebooks
  • Support common agent frameworks: ADK / Camel / LangChain / Autogen

About

An open project initiated by the DeepModeling community focused on constructing intelligent agents for scientific research. It is committed to assembling a reusable and extensible set of Agent-Ready tools, promoting the development and practical deployment of research agents through open collaboration.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages