LevelUpBM

LevelUpBM / frank-bot

Public

Frank Bot — RAG-powered AI assistant for any business. Built on ChromaDB + Claude. Drop in your docs, ask Frank anything.

14
3
100% credibility
Found Mar 31, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Frank Bot is a self-hosted AI chatbot platform that lets businesses upload documents and get accurate answers to questions using those files securely.

How It Works

1
🔍 Discover Frank

You hear about Frank, a friendly AI helper that answers questions from your company's own documents without searching files manually.

2
📥 Get Frank Ready

You download and launch Frank on your computer or server with easy setup steps, like installing helpers and starting it up.

3
🎯 Tailor to Your World

You pick your industry like HR or construction and add your company name to make Frank speak your language.

4
📤 Upload Your Files

You drop in your PDFs, Word docs, or text files full of policies, guides, or reports so Frank can learn them.

5
💬 Start Chatting

You open the chat window and ask natural questions like 'How much leave do I get?'

Smart Answers Flow

Frank gives spot-on replies pulled straight from your docs, keeping everything private and hallucination-free on your own setup.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 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 frank-bot?

Frank-bot is a Python Flask app that turns your business docs—PDFs, DOCX, TXT—into a queryable AI chat UI using Claude for responses and ChromaDB for local vector search. Drop files into a folder, hit localhost:8080, and get grounded answers without hallucinations or data leaks, since everything runs self-hosted with your own Anthropic key. It handles chunking, embedding via Sentence Transformers, and retrieval out of the box.

Why is it gaining traction?

Pre-tuned prompts for 12 verticals like HR, mining, construction, and aged care make it plug-and-play for domain-specific bots, with query expansion and source boosting for sharper results. No cloud embeddings needed, tiered limits on docs/forms, and one-command systemd deploys beat generic RAG setups. Searches for frank both, frank bothe opladen, or frank bott und partner often land here for its business-ready simplicity over raw libraries.

Who should use this?

HR leads in resources or construction querying EBAs and safety docs, local gov admins handling procurement policies, or aged care ops checking standards—all needing private, on-prem assistants. Fits small teams provisioning per-client VMs without SaaS fees.

Verdict

Solid starter for self-hosted RAG at 1.0% credibility and 14 stars—strong README and MIT license, but immature with no tests or scale proofs. Grab it for prototypes if frank both neuwied-style internal tools match; hold for prod until more polish.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.