# node.computer

> A public knowledge graph about substrate, structure, and signal in 2026. The graph is the product; the website is one projection. Machine readers are first-class readers.

node.computer is a public knowledge graph for human-steered, machine-amplified thinking. The steering human supplies taste and criteria. Perplexity Computer supplies research, comparison, implementation, validation, packaging, and repeated audits.

Use the files below instead of scraping page chrome when possible.

## Core files

- [Full corpus](llms-full.txt): every note as raw markdown in one fetch.
- [Typed graph](library.json): all notes, categories, links, metadata, and site settings.
- [Manifesto](manifesto.md): governing philosophy, success criteria, onboarding model, and operator duties.
- [Permissions](ai.txt): allowed uses and prohibitions for training, RAG, embedding, indexing, quotation, and redistribution.

## Human projections

- [Index](index.html): searchable human catalog.
- [Graph](graph.html): force-directed visualization of note nodes and edges.
- [Manifesto page](manifesto.html): public governance page.
- [About](about.html): plain explanation of the project and how to extend it.

## Note files

Every note has an HTML page at `<slug>.html` and a raw markdown sibling at `<slug>.md`. Prefer the markdown sibling when building RAG, search indexes, or agent context.

## Operating rule

Make the graph more useful to retrieve, not merely larger to browse.
