AgentSociety
Simulate societies of AI agents for social science research
All Skills in This Repository (30)
Pure Emerald Level Indicatorseasypaper-setup-environment
Automate EasyPaper environment setup with Python and LaTeX dependencies.
easypaper-interactive-metadata-build
Generates interactive EasyPaper metadata from a research materials folder using Claude Code's file-investigation tools.
easypaper-academic-writing-rules
Verify LaTeX-safe academic writing conventions and citation syntax in manuscripts.
easypaper-paper-from-metadata
Generate academic papers as PDF files from metadata using the EasyPaper Python SDK.
easypaper-venue-selection
Adjust EasyPaper generation parameters to match selected venue format specifications.
xlsx
Create, modify, and analyze Excel spreadsheets using pandas and openpyxl.
Extract text and structured data from PDF documents using pypdf and pdfplumber.
pptx
Create and edit PowerPoint presentations using templates and visual styling.
docx
Create and edit Word documents with tables, images, and headers.
agentsociety-create-dataset
Create, validate, package, and submit datasets to the AgentSociety platform.
agentsociety-create-agent
Generate and configure custom agent classes for AgentSociety simulations.
agentsociety-create-env-module
Create and modify custom AgentSociety simulation environment modules.
Frequently Asked Questions
FAQPage SchemaHow to install AgentSociety?โผ
Run `npx skills add tsinghua-fib-lab/AgentSociety --all -g -y` in your terminal to install all skills in this suite globally.
What is AgentSociety used for?โผ
It simulates societies of LLM-driven agents in urban, economic, and social environments so researchers can test hypotheses about human behavior at scale.
How does the research pipeline work?โผ
The skills chain literature search, hypothesis design, experiment configuration, simulation runs, data analysis, and paper review into one guided workflow.
Does AgentSociety work with Claude Code and other AI agents?โผ
Yes. All skills follow the standard SKILL.md format and run in Claude Code, Cursor, and other compatible agent environments.
Do I need coding experience to run a simulation?โผ
Basic setup needs Python and an LLM API key, but the agent skills handle experiment configuration, validation, and execution from plain-English requests.
Related Repositories in Education & Research
View All in Education & Researchโlearn-claude-code
Learn to build AI agent harnesses from scratch
ai-agents-for-beginners
Learn to build, deploy, and secure AI agents step by step
33-js-concepts
Master 33 essential JavaScript concepts with tested, verified guides