Codeatlas — Data skill for Claude Code
CodeAtlas is a high-performance, local-first intelligent codebase indexer and query engine.
How to install Codeatlas
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open shditz/codeatlas and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Codeatlas does
CodeAtlas is a high-performance, local-first intelligent codebase indexer and query engine. It provides architectural graph analytics, natural language Cypher queries, multi-repository context generation, and seamless IDE integration to help AI agents and developers understand complex codebases.
Alternatives in Data
- SQL Queries — Generate SQL queries from natural language across major dialects 7.8k ★
- Azure Event Hub .net — High-throughput event streaming 1.8k ★
- Claude Code Memory Setup — Up to 71.5x fewer tokens per session on Claude Code with Obsidian + Graphify 947 ★
README
CodeAtlas
**Give AI a map of your codebase.** Local-first context intelligence and architecture platform for AI coding agents.
[](https://github.com/shditz/codeatlas/actions) [](https://opensource.org/licenses/MIT) [](https://www.typescriptlang.org/) [](https://nodejs.org/) [](https://vitest.dev) [](https://github.com/shditz/codeatlas)
💡 What is CodeAtlas?
**CodeAtlas** is a high-performance, privacy-first context engine and code intelligence platform. It parses software repositories locally into an SQLite knowledge graph, retrieves task-relevant files and dependencies with multi-signal semantic search, and packs progressive context tailored for LLMs and AI coding agents.
CodeAtlas is **not** an AI chat assistant—it is the **context layer** that makes all AI coding tools (Cursor, Claude, Antigravity, Copilot, DeepSeek, Trae, Qwen, Kimi, and Grok) substantially faster, more accurate, and token-efficient.
🚀 How It Works
┌─────────────────┐ ┌──────────────────┐ ┌─────────────────────┐ ┌────────────────┐
│ Source Code │ ──> │ Tree-Sitter AST │ ──> │ SQLite Knowledge DB │ ──> │ FTS5 + Graph │
│ (Local-First) │ │ & Symbol Parsing │ │ (.atlas/atlas.db) │ │ Multi-Signal │
└─────────────────┘ └──────────────────┘ └─────────────────────┘ └────────────────┘
Related Skills
Suganthans BigQuery MCP Server
BigQuery MCP server for Claude — query any BigQuery dataset in natural language, with built-in SEO analysis to
Csvql
SQL queries for CSV files. The analytical CSV query engine for AI agents cli: csvql.
Ix Understand
Build a detailed architectural mental model of a system, subsystem, or the whole repo. Graph-first, source rea
Shared Memory
Memory Framework based on the idea of Architectural Decision Records, shared knowledge between AI agents (Clau
Octohub
High-performance LLM proxy: one OpenAI-style API over 20+ providers, with multi-tenant keys, full request logg
Weak Spot Hunter
Analyze Anki review data (exported stats, deck performance) to identify weak spots — cards with low retention,
Related Agents
Brain Query
Query the Brain teaching graph with natural language. Translates to Cypher, synthesizes insights, never expose
Cypher Author
Writes read-only Cypher for natural-language questions given the live graph schema and few-shot examples; also
Graph Retriever
Intelligent, iterative retrieval over the derived graph query layer. Given a maintainer's information need, co