glyphh-ai

Glyphh Code — file-level codebase intelligence via HDC encoding

12
4
100% credibility
Found Mar 27, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Glyphh Code builds a semantic map of code files to supercharge AI coding tools with instant, precise file discovery by meaning.

How It Works

1
🔍 Discover smarter code search

You hear about a tool that lets your AI coding helper instantly find the right files in your project by understanding what they do, not just searching text.

2
📦 Add the tool

You easily add it to your computer with a simple install command, like grabbing a helpful app.

3
🏠 Open your project

You go to your code folder and start the helper's special mode.

4
Build your code map

With one command, it scans your entire project and creates a smart map of every file's purpose and connections, ready in moments.

5
🔗 Connect to your AI

It automatically links to your AI coding assistant, adding new search powers without hassle.

6
💬 Ask natural questions

Now you chat with your AI and say things like 'find auth token code' – it points straight to the perfect files super fast.

🎉 Code smarter and faster

Your AI works quicker and cheaper, spotting risks and connections you might miss, making changes safe and speedy.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 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 glyphh-code?

Glyphh-code is a Python tool for file-level codebase intelligence via HDC encoding. It indexes every source file in your repo as a vector, letting Claude Code query semantically instead of scanning or grepping files. Pip install, enter the glyphh shell, run code init—and you get MCP tools like glyphh_search for NL queries, glyphh_related for blast radius, and glyphh_drift for semantic change scores.

Why is it gaining traction?

Benchmarks on 700-file repos show one glyphh_related call replaces 14-32 tool calls, slashing costs 26-50% and API time 72-79% while delivering the same answers. Pure HDC cosine search means no LLM overhead at build or query time, with incremental git hooks keeping the index fresh. Claude auto-configures hooks and permissions, gating grep until search runs.

Who should use this?

Claude Code users navigating large Python, JS, or Go codebases—think backend devs hunting auth middleware or payment handlers. Teams doing code reviews, measuring commit risk across changed files, or checking edit blast radius before touch. VS Code or CLI workflows where Explore agents burn tokens on endless greps.

Verdict

Promising alpha for Claude Code power users (12 stars, 1.0% credibility score), with strong benchmarks and zero-setup docs outweighing light tests. Eval on a real repo if grep fatigue hits; skip for tiny projects or non-Claude setups.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.