ikchain

Local AI dev assistant powered by Qwen — hybrid RAG, knowledge graph, REST API, and data flywheel

11
1
100% credibility
Found Feb 23, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Fabrik-Codek is a local AI coding assistant that combines your personal knowledge with any open AI model to provide tailored advice through chat, search, and integrations that improve over time.

How It Works

1
🔍 Discover your coding buddy

You hear about Fabrik-Codek, a friendly local helper that gives smart coding advice using your own notes and experiences.

2
📥 Grab it easily

Download the ready-to-go package and place it in your coding folder—no complicated steps needed.

3
🧠 Connect the smart brain

Link it to free AI thinkers on your computer so it can understand and chat about code like a pro.

4
📚 Share your know-how

Point it to your past code notes, decisions, and fixes—it soaks them up to remember your style.

5
💬 Start chatting

Ask questions like 'How do I fix this error?' and get helpful answers pulled from your own wisdom.

6
🔄 It gets smarter

Every chat and fix you share makes it better at helping you, like a teammate who learns on the job.

🚀 Code like a wizard

Now you have a personal coding sidekick that saves time, remembers your tricks, and grows with every project—happy coding!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 11 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Fabrik-Codek?

Fabrik-Codek is a local AI dev assistant that runs entirely on your machine via Ollama models like Qwen, delivering context-aware coding help through a hybrid RAG system combining vector search, knowledge graphs, and full-text lookup. It builds a personal knowledge base from your code changes, sessions, and interactions, feeding that back into every query via CLI commands like `fabrik chat` or `fabrik ask --rag`, plus a FastAPI REST API and MCP server for tool integrations. Python-based, it solves the privacy and lock-in issues of cloud AI by creating a self-improving flywheel of your own dev data.

Why is it gaining traction?

As a local GitHub Copilot alternative, it swaps vendor models for any Ollama option with a config flag, while auto-building profiles of your stack—like preferring FastAPI or Docker setups—and competence maps to tailor responses. The data flywheel captures sessions from tools like Claude Code, turning real work into better future answers without manual effort. Devs dig the offline-first setup, OpenAPI docs at `/docs`, and seamless MCP hooks for Cursor or OpenClaw.

Who should use this?

Backend devs knee-deep in Python/FastAPI/PostgreSQL stacks needing quick RAG-boosted queries during local development with Docker. Solo engineers tired of generic AI responses who want a tool that learns their hexagonal architecture prefs or Terraform habits. Teams running local GitHub Actions runners seeking a private API for custom agents.

Verdict

Promising early experiment with solid docs, 637 passing tests, and MIT license, but only 10 stars and 1.0% credibility signal it's pre-mainstream—test locally before production. Grab it if you need a tunable local dev copilot today.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.