An open-source initiative to build a deeply interconnected, queryable digital canon of human art and thought.
In an age awash with ephemeral content and algorithmic noise, Phantom seeks to carve out a space for enduring works of human creativity and intellect. We face a deluge of information, the fragmentation of cultural memory, and the rise of "AI slop"—content endlessly remixed without grounding in original substance. It's increasingly difficult to connect with the foundational art, literature, and ideas that shape our understanding of ourselves and the world.
Phantom is our response: an ambitious project to build an open-source, canonical library. We are not aiming to catalog everything, but rather to meticulously curate a collection of significant works – the touchstones of art, literature, philosophy, and science – and transform them into a unified, deeply interconnected knowledge graph.
Our core motivations are:
- Combat Information Overload: Provide a focused, high-signal resource amidst the noise, prioritizing depth and significance over sheer volume.
- Preserve & Rediscover: Create a persistent, accessible digital home for canonical works, safeguarding them against cultural amnesia and making them newly discoverable.
- Enable Deep Exploration: Move beyond keyword search. Allow users to explore the semantic relationships, thematic resonances, and intricate connections between works across different domains and eras.
- Foster Meaningful Analysis: Build a structured, vectorized representation of the canon, enabling novel forms of computational analysis, visualization, and understanding.
- Offer an Open Resource: Provide this curated and structured knowledge base as an open-source platform for researchers, students, artists, and the intellectually curious public.
Phantom aims to be more than a digital archive; it aspires to be a dynamic engine for exploring the heights of human expression and thought.
This repository serves as the central coordination point for the entire Phantom system, providing:
- Design system and component library
- Documentation for all system components
- Configuration and standards
- Development environment setup
Phantom is structured as a monorepo using pnpm workspaces and Turborepo for build orchestration. All components are located under the packages/ directory.
The Phantom ecosystem consists of the following components:
- phantom-api: API layer for accessing Phantom services
- phantom-canon: Core content catalog and canon management
- phantom-core: Central hub, documentation, and design system
- phantom-eda: Exploratory data analysis tools
- phantom-editor: Content editing and management interface
- phantom-enrichment: Content enrichment and metadata generation
- phantom-explorer: Interface for exploring the canon
- phantom-folio: User collections and reading lists
- phantom-glacier: Digital asset storage
- phantom-intake: Content ingestion pipeline
- phantom-query: Search and querying infrastructure
- phantom-rd: eBook reader component
- phantom-vector: Vector embedding and semantic search capabilities
- phantomklange: Digital catalogue and user interface
This project uses pnpm for package management and Turborepo for task running.
From monorepo root:
# Clean all packages from root directly
pnpm run cleanFrom monorepo root:
# Install dependencies
pnpm installFrom monorepo root:
pnpm run buildFrom monorepo root, and in one terminal:
pnpm run dev:docsFrom monorepo root, and in one terminal:
pnpm run dev:klangeThis ensures @phantom/core is built first, then @phantom/docs and phantomklange are built using the latest artifacts from @phantom/core.
From monorepo root:
# Build all packages
pnpm run build