agentsociety-analysis

Analyze AgentSociety simulation data with Python and agentsociety2.

1.2k|203|Updated Feb 6, 2025
One-click install
npx skills add https://github.com/tsinghua-fib-lab/AgentSociety --skill agentsociety-analysis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentsociety-analysis
Source: https://github.com/tsinghua-fib-lab/AgentSociety/tree/main/extension/skills/agentsociety-analysis/v1.0.0
Command: npx skills add https://github.com/tsinghua-fib-lab/AgentSociety --skill agentsociety-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agentsociety2, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a comprehensive framework for analyzing simulation data generated by AgentSociety, enabling rigorous interpretation, claim-driven charts, and bilingual reports.

Core Features & Use Cases

  • Interactive Analysis: Facilitates detailed analysis of simulation results through interactive commands.
  • Bilingual Reports: Generates reports in both English and Chinese, enhancing accessibility.
  • Synthesis: Enables synthesis of findings across multiple experiments and hypotheses.
  • Use Case: Imagine you have completed a simulation run with AgentSociety and need to interpret the results. This Skill allows you to explore the data, generate charts, and write detailed analysis reports.

Quick Start

To analyze a simulation run, load the context with the ags.py analysis load-context command, then proceed with data exploration, claims recording, and report generation.

Frequently Asked Questions about agentsociety-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze AgentSociety simulation data and generate interactive reports?▼

You can analyze AgentSociety simulation data by loading the context first, then performing interactive data exploration, generating claim-driven charts, and creating bilingual reports to synthesize findings across multiple experiments.

What is the best way to visualize findings across multiple AgentSociety simulation experiments?▼

Synthesize findings across multiple AgentSociety experiments by recording claims during interactive data exploration, then automatically generating claim-driven charts and bilingual reports to summarize the results.

Do I need Python and the agentsociety2 library to run simulation analysis workflows?▼

Yes, the simulation analysis workflow requires Python and the agentsociety2 library installed, as these dependencies provide the necessary functions for data manipulation, interactive exploration, and visualization.

Can I generate simulation analysis reports in both English and Chinese?▼

Yes, the framework supports bilingual report creation, allowing you to generate detailed simulation analysis reports in both English and Chinese to enhance accessibility for diverse audiences.

How does claim-driven chart generation work for simulation data analysis?▼

Claim-driven chart generation works by linking recorded analytical claims to simulation data, automatically producing targeted visualizations that validate hypotheses during interactive data exploration.