IOchair

IOchair / SQL-ManyThing

Public

Turn any source tree into a local SQLite database. FTS5 trigram search across 89K files in seconds. One file. No server. No network.

25
2
85% credibility
Found May 26, 2026 at 25 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

SQL-ManyThing is a local search tool that transforms any folder of source code into a fast, searchable SQLite database. It works in three stages: first, it indexes all your files into a single database with full-text search capabilities. Second, it can optionally enrich your search with information about code structures like functions, classes, and their relationships. Third, it keeps a history of every search you run so you can revisit and reuse past queries. The tool is designed to help developers and AI assistants find specific code quickly without reading entire files or relying on network services.

How It Works

1
💡 You have a big codebase to search

You have thousands of files to dig through and regular search tools are too slow or clumsy.

2
🔍 You turn your whole project into a searchable database

One simple command indexes every file in your project into a single database that lives on your computer.

3
⚡ Your project is now lightning-fast to search

Instead of reading files one by one, you ask questions and get answers in seconds with full-text search.

4
🎯 You optionally add extra knowledge about your code

If you want deeper understanding, you can add information about function names, class structures, and code relationships.

5
Choose your search style
📄
Find files by content

Search for text that appears anywhere in your code

🏗️
Find by code structure

Look up specific functions, classes, or symbols by name

🎉 You find exactly what you need

Your search returns precise results with evidence from the source, and the system remembers your queries for next time.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 25 to 25 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 SQL-ManyThing?

SQL-ManyThing is a Python tool that converts any codebase into a local SQLite database with full-text search powered by FTS5 trigrams. You point it at a directory, it indexes every file, and you query the result with SQL. The project ships as standalone scripts organized in three phases: build the index, optionally enrich it with symbol and graph data, then track every query you run. The stress test claims 89K files indexed in seconds against an Unreal Engine install, producing a single 3GB database that stays entirely local.

Why is it gaining traction?

The pitch is simple: grep is linear scan, LSP is language-locked, cloud RAG needs a network. SQL-ManyThing is offline-first, language-agnostic, and uses bounded substring extraction so you never blow your context window reading whole files. For developers building AI agents that search code, this turns a codebase into a queryable knowledge base with auditable SQL history. The query tracing phase records every search, letting agents learn from past sessions. It is the kind of tool that feels obvious in hindsight but nobody had packaged this way before.

Who should use this?

AI agent developers building code-understanding tools will get the most value. If you are writing a system that needs to search across large, heterogeneous codebases without shipping data to a remote service, this fits. Researchers working with Unreal Engine or similarly large C++ codebases might find the UHT reflection enrichment useful. For small projects where grep works fine, the overhead is not worth it.

Verdict

The core idea is solid and the FTS5 approach is the right primitive for local code search, but the project is early: 25 stars, thin test coverage, and documentation that assumes familiarity with the author's workflow. The 0.85% credibility score reflects that this is a niche tool from an unknown author, not a battle-tested library. Try it on a personal project to see if the workflow fits, but do not bet a production system on it yet.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.