jyshnkr/repo-indexer

3 stars · Last commit 2026-02-26

A Claude skill that indexes codebases with minimal context‑window overhead using tiered memory.

PluginWorkflow & Automation TDD & Testing Debugging Design & UI DevOps & Infra Productivity Data & ML

README preview

# repo-indexer

**Index any codebase for persistent Claude context — minimal token overhead between sessions (~500 tokens for CLAUDE.md boot).**

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
[![Version](https://img.shields.io/badge/version-0.0.3-blue.svg)](.claude-plugin/plugin.json)
[![Claude Code Plugin](https://img.shields.io/badge/Claude%20Code-Plugin-orange.svg)](https://claude.ai/plugins)

---

## The Problem

Every new Claude session starts blank. You re-explain your architecture, your conventions, your stack — burning hundreds of tokens just to get Claude up to speed. For large codebases, this context tax is constant and expensive.

## The Solution

repo-indexer runs a structured 6-phase analysis of your codebase and writes the results into a **tiered memory system** that scales across sessions with near-zero overhead:

```
L0: Claude Native Memory  → repo roster, patterns (~100 tokens, always present)

View full repository on GitHub →