AI-ready data stacks aren’t future talk—they’re live. Agentset bundles a MIT-licensed open-source RAG with a vector DB, embeddings, and an API, built around big-scale work with Usul.ai and early bets on LangChain and LlamaIndex [1]. On the publishing side, WP-MCP demonstrates AI-friendly WordPress workflows via a lean MCP server [2].
Agentset snapshot Agentset is an MIT-licensed open-source RAG with a vector DB, embeddings, and an API built in. It covers 22 file formats, agentic search, deep research, citations, and a UI out of the box [1]. The project started with LangChain and LlamaIndex to prototype at scale before packaging for production [1].
WP-MCP + WordPress workflows WP-MCP exposes WordPress management to AI clients. It connects directly to the WordPress database through wp-node (a TypeScript library) that mirrors core CRUD operations [2]. It can create and update posts, manage users and categories, and move drafts through review-to-publish flows. Access runs via STDIO for local clients and Streamable HTTP for remote setups [2].
What AI-ready today means: - Retrieval with embeddings and vector search built in, as demonstrated by Agentset [1]. - Built-in API access to power AI agents and tooling [1]. - AI-enabled content workflows, from draft to publish, illustrated by WordPress publishing via MCP [2].
Closing thought: AI-ready means modular, embeddable retrieval plus end-to-end content workflows—watch open-source RAG stacks merged with CMS-driven publishing grow in 2025.
References
Show HN: Agentset – Open-source RAG with vector DB, embeddings, and API built-in
Show HN: Agentset opensource RAG stack with vector DB, embeddings, multi-format support, and a built-in UI for production-ready deployment today.
View sourceShow HN: WP-MCP – An MCP server to control WordPress
Announces WP-MCP, an MCP server exposing WordPress DB tools to AI clients for content publishing without PHP and local/remote access.
View source