This project provides an open-source political technology platform designed for the AI agent era, aiming to democratize access to political decision-making processes. It serves citizens, researchers, and AI agents by offering transparent, non-partisan, and non-corporate infrastructure to monitor political activities, analyze policies, and understand governmental functions, thereby empowering individuals in the digital age.

How It Works

The platform adopts an "agent-first" and "API-first" design philosophy, prioritizing machine readability and direct AI agent interaction. Its architecture is a monorepo built with Next.js 15, React 19, and PostgreSQL via Supabase, managed by pnpm. Key innovations include the BroadListening app, which models public opinion as a digital ecosystem using Argument Mining and Stigmergy for self-organizing consensus, and a visually rich portal leveraging WebGPU for fluid data visualizations.

Quick Start & Requirements

A one-line installation script is available: curl -fsSL https://ojpp.dev/install | sh. This requires Docker and Git; Node.js and pnpm are installed automatically. Manual setup involves cloning the repository, setting up PostgreSQL, and running ingestion scripts. Links to official quick-start guides and documentation are implicitly available through the project's structure and README.

Highlighted Details

• Seven Integrated Apps: Includes MoneyGlass (political finance visualization), PolicyDiff (policy comparison), ParliScope (parliamentary data API), SeatMap (parliamentary seat visualization), CultureScope (cultural policy analysis), SocialGuard (social security overview), and BroadListening (opinion ecosystem beta).
• Rich User Experience: Features smooth animations and scrolling powered by Motion (v11+) and Lenis, alongside WebGPU for dynamic dashboard visualizations.
• AI Agent Integration: Designed for AI agents with API-first principles, machine-readable data formats, and planned support for agent authentication and the Model Context Protocol (MCP). The Entire framework enables AI session persistence and attribution.
• Data Ingestion: Pipelines ingest real-world data from various Japanese government sources, including political funding reports, parliamentary records, cultural budgets, and social security data.

Maintenance & Community

The project welcomes contributions from both humans and AI agents, with clear guidelines provided in CONTRIBUTING.md. While specific community channels like Discord or Slack are not explicitly listed, the project structure and roadmap indicate active development and a focus on open collaboration.

Licensing & Compatibility

The platform is licensed under AGPL-3.0-or-later. This strong copyleft license requires that any derivative works or modifications distributed must also be made available under the same license, which may impose restrictions on integration with closed-source commercial products.

Limitations & Caveats

The BroadListening application is currently in beta. Features such as user authentication, real-time notifications, and multi-language support are planned for future releases (v0.2). Some data, like party stances on cultural and social security policies, is noted as demo data, while other datasets are derived from public sources.