ind-igo

ind-igo / cx

Public

Semantic code navigation for AI agents

89
2
100% credibility
Found Mar 29, 2026 at 89 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

cx is a tool that scans code projects to provide AI agents with efficient summaries, symbol searches, definitions, and usage locations without needing to read entire files.

How It Works

1
🔍 Discover cx

You hear about a smart helper that lets AI assistants quickly peek into your code projects without reading every single line.

2
📥 Install cx

You grab the tool onto your computer using a simple download script or your usual software installer.

3
📁 Open your project

You go to your code folder, and cx automatically builds a quick map of all the files and their key parts.

4
🧠 Guide your AI

You share a ready-made tip sheet with your AI buddy, teaching it clever ways to ask cx for code summaries and lookups.

5
🔎 Get code insights

Your AI now asks cx for a file overview, function details, or where something is used, getting answers in seconds.

🚀 AI works smarter

Your AI navigates the codebase lightning-fast, saving tons of time and effort while tackling tasks efficiently.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 89 to 89 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 cx?

cx is a Rust CLI for semantic code navigation in codebases, built for AI agents like Claude or Copilot. It indexes files with Tree-sitter grammars (rust, typescript, python, go, and more via `cx lang add`), delivering `cx overview file.rs` for symbol tables, `cx symbols --kind fn` for project-wide searches, `cx definition --name foo` for function bodies, and `cx references --name foo` for usages. Cuts AI token waste on full file reads by escalating from cheap summaries (~70-200 tokens) to precise extracts.

Why is it gaining traction?

Beats LSP daemons (no 1-2GB RAM, instant queries) and grep/ctags (semantic parsing over regex). Agents follow a "cost ladder" via `cx skill` prompts, slashing 58% of reads and 40-55% tokens per session per their analysis. If hunting github cx, cx explorer github, or github semantic kernel, this delivers aruba cx github-style precision without the hardware—far from cx_freeze or github cx patcher.

Who should use this?

AI coding agent users on Rust/TS/Python/Go projects, like backend devs chaining file explorations in Claude Code or frontend teams querying React components in Cursor. Ideal for mid-size repos where agents re-read files or hunt symbols amid cx30/CX5 Mazda distractions in searches.

Verdict

Promising for AI workflows at 56 stars and 1.0% credibility—docs shine, tests cover edge cases, but watch for lang gaps. Install via Cargo, gitignore the DB, and test on your repo; maturity lags big tools but hooks agent efficiency fast. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.