core  by mx-space

AI-powered CMS core for blogs and creator sites

Created 4 years ago
491 stars

Top 63.1% on SourcePulse

GitHubView on GitHub
Project Summary

AI-powered CMS core for personal blogs and creator websites, offering integrated AI content workflows such as summarization, translation, moderation, and writing assistance. It targets independent developers and content creators seeking a self-hostable, extensible backend with multi-provider LLM support, providing a robust infrastructure for content management and AI-enhanced publishing.

How It Works

MX Space Core is a headless CMS backend built on NestJS and TypeScript, designed to power personal blogs and creator websites. It integrates a comprehensive AI content workflow, enabling features like AI-driven content generation, summarization, translation, and moderation. The architecture supports multiple LLM providers (OpenAI, Anthropic, OpenRouter, compatible) and handles AI responses through streaming and task-based processing, facilitating seamless integration into content creation pipelines. Its monorepo structure separates core services, API clients, and webhook SDKs.

Quick Start & Requirements

  • Environment: Node.js >= 22, pnpm (recommended via Corepack).
  • Install: pnpm i
  • Run: pnpm dev (development), pnpm build (build), pnpm bundle (production bundle).
  • Other Commands: pnpm test, pnpm lint, pnpm typecheck.

Highlighted Details

  • AI Content Workflow: Features AI summarization (generation/auto), multi-language translation (auto), comment moderation (spam detection), and writing assistance (title/slug generation).
  • Multi-Provider LLM Support: Integrates with OpenAI, OpenAI-compatible endpoints, Anthropic, and OpenRouter.
  • Headless API: Provides a robust, NestJS-based API for flexible frontend and client integrations.
  • Self-Hosted & Extensible: Tailored for independent developers and self-hosting, offering a backend kernel for customization.
  • Comprehensive CMS: Includes full content management capabilities for articles, categories, tags, pages, drafts, subscriptions, and comments.

Maintenance & Community

No specific details regarding contributors, sponsorships, or community channels were provided in the README.

Licensing & Compatibility

  • License: Files within the apps/ directory are licensed under GNU Affero General Public License v3.0 (AGPLv3) with Additional Terms. Other parts of the repository use the MIT License.
  • Compatibility: The AGPLv3 license imposes strong copyleft requirements, potentially affecting commercial or closed-source integrations. The MIT licensed components offer permissive usage.

Limitations & Caveats

Upgrading from v9 to v10 involves a breaking change in the authentication system. The dual licensing structure (AGPLv3 for core app, MIT for packages) requires careful consideration for integration, particularly regarding the AGPLv3's copyleft provisions.

Health Check
Last Commit

19 hours ago

Responsiveness

Inactive

Pull Requests (30d)
34
Issues (30d)
6
Star History
9 stars in the last 30 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), Harrison Chase Harrison Chase(Founder of LangChain), and
1 more.

gpt-newspaper by rotemweiss57

0%
1k
Autonomous agent for personalized newspapers
Created 2 years ago
Updated 1 year ago
Feedback? Help us improve.