RAG engine for deep document understanding

.. PLUS: Add Semantic Search to Claude Code

Author

Sumanth P
August 01, 2025

In today’s newsletter:

  • Code Context: Add semantic search to Claude Code

  • RAGFlow: A RAG engine for deep document understanding

  • FireGEO: Open-source tool to monitor your site’s AI search visibility

Reading time: 3 minutes.

Coding agents like Claude Code, Gemini CLI, Cursor, and others share a major limitation: they forget everything between sessions.

Without memory, they can’t retain project context, past fixes, or key decisions. What they need is a persistent memory layer to store and recall context.

Code Context is an MCP plugin that adds semantic code search to Claude Code and other AI coding agents, giving them deep, project-wide context.

Unlike traditional coding assistants limited by context windows, Code Context uses semantic search powered by Zilliz Cloud to find and understand relevant code from your entire project using natural language, not just keywords.

This gives your assistant deep context awareness for better code generation.

Key Features:

  • Semantic Code Search: Ask questions like “find functions handling user authentication” and get instant, relevant results.

  • Context-Aware: Understands relationships across large, complex codebases.

  • Incremental Indexing: Quickly re-index only changed files using Merkle trees.

  • AST-Based Chunking: Breaks code into meaningful chunks using Abstract Syntax Trees.

  • Scalable: Handles huge codebases with Zilliz Cloud.

  • Customizable: Configure extensions, ignore patterns, and embedding models.

It’s 100% open source

A new open-source RAG engine built to handle complex data formats with ease.

RAGFlow makes it easier to extract insights from real-world documents PDFs, spreadsheets, scans, images by deeply understanding their structure.

Key Features:

  • Template-based chunking for document-aware parsing

  • Source-cited answers with full traceability

  • Combines vector + full-text search

  • Supports multimodal queries and code-executing agents

Perfect for developers building robust RAG systems beyond simple text files.

It’s 100% Open Source

FireGEO is a full-stack, open-source starter template for building Generative Engine Optimization (GEO) apps. It’s designed for developers who want to launch production-ready SaaS products quickly without rebuilding the basics.

Why GEO?

Search is changing. Users are discovering brands and products through AI engines like ChatGPT, Perplexity, and Claude. GEO (Generative Engine Optimization) is about monitoring and improving your visibility across these new discovery channels.

FireGEO gives you the tools to build apps that track, analyze, and act on this data.

What’s inside:

  • Auth, billing, and database setup with Better Auth, Autumn + Stripe, and PostgreSQL (via Drizzle)

  • Firecrawl integration for scraping and analyzing brand mentions with LLMs like OpenAI, Anthropic, and Perplexity

  • Modern stack with Next.js, TypeScript, and Tailwind for a clean, production-ready codebase

  • Fully open source under the MIT license so you can own and extend everything

That’s a Wrap

That’s all for today. Thank you for reading today’s edition. See you in the next issue with more AI Engineering insights.

PS: We curate this AI Engineering content for free, and your support means everything. If you find value in what you read, consider sharing it with a friend or two.

Your feedback is valuable: If there’s a topic you’re stuck on or curious about, reply to this email. We’re building this for you, and your feedback helps shape what we send.

WORK WITH US

Looking to promote your company, product, or service to 120K+ AI developers? Get in touch today by replying to this email.