AI Content Knowledge Base is a reference architecture and template for a review-first, AI-native knowledge base using Markdown, YAML sidecars, and labeled vault sections. It emphasizes separate provenance layers, a staged workflow, and typed relationships, with Obsidian and Codex workflows mentioned.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (2 so far).
What it is
AI Content Knowledge Base is a reference architecture and working template for a review-first, AI-native knowledge base intended for researchers, writers, teachers, and publishers. It combines Markdown, Codex workflows, Obsidian compatibility, and typed YAML relationships to organize content across provenance layers.
How it works
The repository defines a multi-layer vault structure with named folders: raw/, sources/, products/, wiki/, .kb/links/, and .kb/staging/. Ingested content from raw/ and sources/ flows into .kb/staging/ for AI-generated or drafted material, which is then reviewed by humans before moving into wiki/ and products/. Typed relationships are stored under .kb/links/, and provenance and versioning are enforced by keeping originals separate from published outputs. The workflow includes considering gatekeeping steps where AI output passes through staging and human review prior to publication.
Getting started
Quickstart steps:
git clone https://github.com/mrbear1024/ai-content-kb.git
cd ai-content-kb
No application is required to read the vault; it can be read as plain Markdown or opened as an Obsidian vault. The repository provides workflows for Codex integration and links AGENTS.md for Codex guidance.
Recent releases
RELEASES (latest 0):
- none
Traction
Stars: 105 Forks: 6 Open issues: 0
Behind the repo
Maintained by mrbear1024 and contributors. The project is described as a reference architecture and working template rather than a hosted app, with Obsidian and Codex as optional interfaces.
Caveats
License: MIT Created: 2026-07-11 Last push: 2026-07-11