Coraline

Fast, Local Code Intelligence for AI Assistants

Semantic code indexing • Symbol-level editing • 100% Rust • MCP Server

What is Coraline?

Coraline is a Rust implementation that combines the best ideas from two powerful coding tools:

CodeGraph - Semantic code knowledge graphs for efficient AI exploration
Serena - Symbol-level code understanding and editing tools

Built from the ground up in Rust, Coraline provides:

Native performance - Fast indexing and queries without Node.js overhead
Semantic search - Find code by meaning using vector embeddings
Symbol-level tools - IDE-like precision for AI assistants
100% local - All processing happens on your machine
MCP integration - Works with Claude Desktop, Claude Code, and other MCP clients

Key Features

From CodeGraph

Semantic Knowledge Graph - Pre-indexed symbol relationships and call graphs
Multi-language Support - 33 languages including TypeScript, Rust, Python, Go, C#, Java, C/C++, Ruby, Bash, PHP, Swift, Kotlin, Markdown, TOML
Vector Embeddings - Semantic code search using local ONNX models
Impact Analysis - Understand what code changes will affect
Git Integration - Auto-sync on commits to keep index fresh

From Serena

Symbol-level Operations - Find, read, and edit code at the function/class/method level
Reference Resolution - Find all references to a symbol across the codebase
Precise Editing - Insert before/after symbols, replace symbol bodies
Project Memories - Persistent knowledge storage for project context
Smart Context Building - Gather relevant code for AI assistants efficiently

Installation

From crates.io (recommended)

cargo install coraline

This builds and installs the coraline binary to your Cargo bin directory and adds it to your PATH.

Platform	Install location
Linux / macOS	`~/.cargo/bin/coraline`
Windows	`%USERPROFILE%\.cargo\bin\coraline.exe`

Windows note: After installation completes the binary is at %USERPROFILE%\.cargo\bin\coraline.exe. This directory is automatically added to PATH by the Rust installer (rustup). Open a new terminal and run coraline --version to confirm it's working. If you see "command not found", add %USERPROFILE%\.cargo\bin to your PATH manually via System Properties → Environment Variables.

Pre-built binaries

Download the latest release archive for your platform from the Releases page:

Platform	Archive
Linux x86_64	`coraline-linux-x86_64.tar.gz`
Linux ARM64	`coraline-linux-aarch64.tar.gz`
macOS ARM64 (Apple Silicon)	`coraline-macos-aarch64.tar.gz`
Windows x86_64	`coraline-windows-x86_64.exe.zip`

Extract the archive and move the binary somewhere on your PATH:

# Linux / macOS
tar xzf coraline-*.tar.gz
sudo mv coraline /usr/local/bin/

# Windows (PowerShell)
Expand-Archive coraline-windows-x86_64.exe.zip .
# Move coraline.exe to a directory on your PATH, e.g.:
Move-Item coraline.exe "$env:USERPROFILE\.cargo\bin\"

Build from Source

git clone https://github.com/greysquirr3l/coraline.git
cd coraline
cargo install --path crates/coraline --force

Semantic Search / LLM Embeddings

Semantic search is included by default — cargo install coraline bundles ONNX/vector-embedding support (via the embeddings feature). No extra flags required.

After running coraline init, you will be prompted to download the embedding model (~137 MB) if stdin is a TTY:

Download embedding model for semantic search? (~137 MB) [Y/n]

You can always download or re-download the model manually:

# Download the int8-quantised model (~137 MB) from HuggingFace
coraline model download

# Generate embeddings for the indexed project
coraline embed

# Skip automatic pre-embed sync check (advanced)
coraline embed --skip-sync

# Combine both steps
coraline embed --download

# Check which model files are present
coraline model status

Models are stored per-project in .coraline/models/. If no model is present, coraline_semantic_search is simply not registered as an MCP tool — all other tools remain fully functional.

Pre-built Binary Feature Matrix

Build	Embeddings	ONNX Runtime
`coraline-linux-x86_64`	Full	Bundled
`coraline-macos-aarch64`	Full	Bundled
`coraline-windows-x86_64`	Full	Bundled
`coraline-linux-aarch64`	Dynamic	Requires `libonnxruntime`
`coraline-linux-x86_64-musl`	Dynamic	Requires `libonnxruntime`
`coraline-linux-aarch64-musl`	Dynamic	Requires `libonnxruntime`

Dynamic builds compile full embedding support but load ONNX Runtime at runtime. Install libonnxruntime via your package manager or from ONNX Runtime releases. If the library is not found, embeddings gracefully degrade — all other tools remain functional.

Older Linux / HPC systems (glibc issues)

If the embeddings feature fails to compile due to glibc incompatibility (e.g., Rocky Linux, CentOS, HPC nodes), use embeddings-dynamic instead — it links against a system-installed libonnxruntime.so at runtime:

cargo install coraline --no-default-features --features embeddings-dynamic

You must have ONNX Runtime installed and on your library path (LD_LIBRARY_PATH or /usr/local/lib).

Alternatively, download a musl static binary from the Releases page — zero glibc dependency (requires libonnxruntime for embeddings).

Quick Start

1. Initialize a Project

cd your-project
coraline init

This creates a .coraline/ directory with:

SQLite database for the code graph
Configuration file
Project memories

2. Index Your Code

coraline index

Coraline will:

Parse source files using tree-sitter
Extract symbols (functions, classes, methods, types)
Build the call graph and reference map

Semantic search: run coraline embed after indexing to generate vector embeddings (model download is prompted automatically on coraline init).

3. Use as MCP Server

Configure your MCP client to use Coraline:

MCP protocol: negotiates to 2025-11-25 (falls back to 2024-11-05 for older clients)
Lifecycle: clients should send notifications/initialized after initialize before normal requests
tools/list supports cursor pagination (cursor / nextCursor)

For Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "coraline": {
      "command": "/path/to/coraline",
      "args": ["serve", "--mcp"],
      "env": {}
    }
  }
}

For Claude Code (.claude/mcp.json in your workspace):

{
  "mcpServers": {
    "coraline": {
      "command": "coraline",
      "args": ["serve", "--mcp"]
    }
  }
}

CLI Usage

coraline init [path]              # Initialize project
coraline index [path]             # Build code graph
coraline sync [path]              # Incremental update (git-diff based)
coraline status [path]            # Show project status and paths
coraline stats [path]             # Show index statistics
coraline query <search>           # Search symbols
coraline context <task>           # Build AI context
coraline callers <node-id>        # Find what calls a symbol
coraline callees <node-id>        # Find what a symbol calls
coraline impact <node-id>         # Analyze change impact
coraline config [--set key=val]   # Read or update configuration
coraline hooks install|remove     # Manage git post-commit hook
coraline serve --mcp              # Start MCP server

See the published CLI reference: https://greysquirr3l.github.io/coraline/cli-reference.html.

MCP Tools

When running as an MCP server, Coraline exposes 27 standard tools prefixed with coraline_, plus optional coraline_semantic_search when the embedding model has been downloaded (for 28 total tools in that configuration; see Semantic Search). See the published MCP tools reference: https://greysquirr3l.github.io/coraline/mcp-tools.html.

coraline_semantic_search also performs periodic freshness maintenance: it checks index staleness on an interval, auto-runs incremental sync when needed, and refreshes stale embeddings before returning results.

Graph Tools

Tool	Description
`coraline_search`	Find symbols by name or pattern
`coraline_callers`	Find what calls a symbol
`coraline_callees`	Find what a symbol calls
`coraline_impact`	Analyze change impact radius
`coraline_dependencies`	Outgoing dependency graph from a node
`coraline_dependents`	Incoming dependency graph — what depends on a node
`coraline_path`	Find a path between two nodes
`coraline_stats`	Detailed statistics by language, kind, and edge type
`coraline_find_symbol`	Find symbols with rich metadata + optional body
`coraline_get_symbols_overview`	List all symbols in a file
`coraline_find_references`	Find all references to a symbol
`coraline_node`	Get full node details and source code

Context Tool

Tool	Description
`coraline_context`	Build structured context for an AI task

Audit Tool

Tool	Description
`coraline_audit_docs`	Audit Markdown docs for stale references and undocumented exports

File & Config Tools

Tool	Description
`coraline_read_file`	Read file contents
`coraline_list_dir`	List directory contents
`coraline_get_file_nodes`	Get all indexed nodes in a file
`coraline_status`	Show project index statistics
`coraline_sync`	Trigger incremental index sync
`coraline_get_config`	Read project configuration
`coraline_update_config`	Update a config value
`coraline_semantic_search`	Vector similarity search (requires embeddings)

Memory Tools

Tool	Description
`coraline_write_memory`	Write or update a project memory
`coraline_read_memory`	Retrieve a stored memory
`coraline_list_memories`	List all memories
`coraline_delete_memory`	Remove a memory
`coraline_edit_memory`	Edit memory via literal or regex replace

Architecture

Diagram source: docs/diagrams/readme-architecture-overview.mmd

Data Flow

Parsing: tree-sitter extracts AST and symbols
Storage: Nodes and edges stored in SQLite
Resolution: References resolved to definitions
Embeddings: Symbols embedded for semantic search (ONNX)
Query: Graph traversal + vector similarity
Serve: Results returned via MCP protocol

Supported Languages

Coraline uses tree-sitter for fast, accurate code parsing. Current support:

Fully Implemented

Language	Parser	Status	Notes
TypeScript	tree-sitter-typescript	Full	Functions, classes, methods, interfaces
JavaScript	tree-sitter-javascript	Full	ES6+, JSX support
Rust	tree-sitter-rust	Full	Full symbol extraction
Python	tree-sitter-python	Full	Classes, functions, methods
Go	tree-sitter-go	Full	Packages, functions, structs
Java	tree-sitter-java	Full	Classes, methods, interfaces
C	tree-sitter-c	Full	Functions, structs, typedefs
C++	tree-sitter-cpp	Full	Classes, templates, namespaces
C# (.NET)	tree-sitter-c-sharp	Full	ASP.NET Core, Blazor, .razor files
Ruby	tree-sitter-ruby	Full	Classes, modules, methods
Bash	tree-sitter-bash	Full	Shell scripts, functions
Dart	tree-sitter-dart	Full	Classes, functions, widgets
Elixir	tree-sitter-elixir	Full	Modules, functions, macros
Elm	tree-sitter-elm	Full	Functions, types, modules
Erlang	tree-sitter-erlang	Full	Modules, functions
Fortran	tree-sitter-fortran	Full	Subroutines, functions, modules
Groovy	tree-sitter-groovy	Full	Classes, methods, closures
Haskell	tree-sitter-haskell	Full	Functions, types, typeclasses
Julia	tree-sitter-julia	Full	Functions, types, modules
Kotlin	tree-sitter-kotlin-ng	Full	Classes, functions, objects
Lua	tree-sitter-lua	Full	Functions, tables, modules
Markdown	tree-sitter-markdown-fork	Full	Documents, headings, lists
MATLAB	tree-sitter-matlab	Full	Functions, scripts
Nix	tree-sitter-nix	Full	Derivations, functions
Perl	tree-sitter-perl	Full	Packages, subroutines
PHP	tree-sitter-php	Full	Classes, functions, methods
PowerShell	tree-sitter-powershell	Full	Functions, cmdlets, scripts
R	tree-sitter-r	Full	Functions, scripts
Scala	tree-sitter-scala	Full	Classes, objects, traits
Swift	tree-sitter-swift	Full	Classes, structs, functions
TOML	tree-sitter-toml-ng	Full	Configuration, tables, keys
YAML	tree-sitter-yaml	Full	Structure, mappings
Zig	tree-sitter-zig	Full	Functions, structs

In Progress

Language	Status	Notes
Liquid	Pending	Parser API compatibility issue

Note: Language support requires tree-sitter grammar integration. Some parsers require older tree-sitter versions and will be added when updated parsers are available.

Configuration

Configuration lives in .coraline/config.toml. A commented template is created by coraline init.

[indexing]
max_file_size = 1048576   # 1 MB
batch_size    = 100

[context]
max_nodes          = 20
max_code_blocks    = 5
max_code_block_size = 1500
traversal_depth    = 1

[sync]
git_hooks_enabled = true

[vectors]
enabled = false   # ONNX embedding support (pending stable ort 2.0)

See the published configuration docs: https://greysquirr3l.github.io/coraline/configuration.html.

Testing

# Run all tests
cargo test --all-features

# Run with output
cargo test --all-features -- --nocapture

# Run specific integration test file
cargo test --test context_test

Current status: 38/38 tests passing See the published development guide: https://greysquirr3l.github.io/coraline/development.html.

Documentation

Primary docs are published on GitHub Pages:

Docs homepage: https://greysquirr3l.github.io/coraline/
Source: docs/book/
Workflow: .github/workflows/docs-pages.yml
Published URL: https://greysquirr3l.github.io/coraline/

Build locally:

mdbook build docs/book

Document	Description
https://greysquirr3l.github.io/coraline/	mdBook home and navigation
https://greysquirr3l.github.io/coraline/architecture.html	System design and data model
https://greysquirr3l.github.io/coraline/mcp-tools.html	Complete MCP tools reference
https://greysquirr3l.github.io/coraline/cli-reference.html	All CLI commands and flags
https://greysquirr3l.github.io/coraline/configuration.html	Configuration guide
https://greysquirr3l.github.io/coraline/development.html	Build, test, and contribute

Contributing

Contributions welcome! See https://greysquirr3l.github.io/coraline/development.html for build setup, coding style, and how to add new tools and language parsers.

License

Licensed under either of:

Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

Acknowledgements

Coraline is inspired by and built upon ideas from:

CodeGraph by Colby McHenry - Semantic code graphs for AI
Serena by Oraios AI - Symbol-level code intelligence
tree-sitter - Fast, incremental parsing library

References

CodeGraph: https://github.com/colbymchenry/codegraph
Serena: https://github.com/oraios/serena
MCP: https://modelcontextprotocol.io/
tree-sitter: https://tree-sitter.github.io/

Built with Rust for the AI coding community

Report Bug · Request Feature

Name		Name	Last commit message	Last commit date
Latest commit History 133 Commits
.cargo		.cargo
.github		.github
assets/img		assets/img
crates		crates
docs		docs
.gitignore		.gitignore
.markdownlint.json		.markdownlint.json
CHANGELOG.md		CHANGELOG.md
CLAUDE.md		CLAUDE.md
Cargo.toml		Cargo.toml
LICENSE-APACHE		LICENSE-APACHE
LICENSE-MIT		LICENSE-MIT
README.md		README.md
SECURITY.md		SECURITY.md
deny.toml		deny.toml

Folders and files

Latest commit

History

Repository files navigation

Coraline

Fast, Local Code Intelligence for AI Assistants

What is Coraline?

Key Features

From CodeGraph

From Serena

Installation

From crates.io (recommended)

Pre-built binaries

Build from Source

Semantic Search / LLM Embeddings

Pre-built Binary Feature Matrix

Older Linux / HPC systems (glibc issues)

Quick Start

1. Initialize a Project

2. Index Your Code

3. Use as MCP Server

CLI Usage

MCP Tools

Graph Tools

Context Tool

Audit Tool

File & Config Tools

Memory Tools

Architecture

Data Flow

Supported Languages

Fully Implemented

In Progress

Configuration

Testing

Documentation

Contributing

License

Acknowledgements

References

About

Topics

Resources

License

Licenses found

Security policy

Uh oh!

Stars

Watchers

Forks

Releases 22

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages