CodeGraph logo

CodeGraph

Organization
Synaptic-Graph

Auditable code intelligence for AI agents - map dependencies, predict change impact, select the right tests, and verify refactors before they ship.

PublisherSynaptic-Graph
RepositorySynaptic
LanguageRust
Forks
1
Stars
28
Available tools
0
Transport typestdio
Categories
LicenseAGPL-3.0
Links
  • Connect tools to AI workflows

    CodeGraph exposes MCP capabilities that can be used by compatible AI clients and agents.

  • 0 available tools

    Browse the callable actions below, including names and descriptions when provided by the server.

  • Ready-to-copy setup

    Use the installation snippets to configure this server in your preferred MCP client.

  • Open source signals

    28 stars and 1 forks from the linked repository.

Synaptic

Give your coding assistant a map of the codebase, not a pile of files.

Synaptic maps a repository once so you and your coding assistant do not have to rediscover it every session. Ask how a system works, what depends on a symbol, or what a change may break. The answers come from a persistent, source-grounded knowledge graph, available from the CLI or through MCP. Synaptic can also remember past changes and plan small, reviewable API and dependency repairs.

Code extraction is local and deterministic. Synaptic ships as a single Rust binary with no runtime, database, account, or API key required.

Documentation | Quickstart | Benchmarks | Discord

Get started

Install the latest checksummed release:

sh
# macOS / Linux
curl --proto '=https' --tlsv1.2 -LsSf https://raw.githubusercontent.com/ColinVaughn/Synaptic/main/install.sh | sh

# Windows PowerShell
irm https://raw.githubusercontent.com/ColinVaughn/Synaptic/main/install.ps1 | iex

Then run Synaptic from a repository:

sh
synaptic extract .                         # build synaptic-out/graph.json
synaptic query "how does authentication work?"
synaptic affected parse_config             # what could this change break?

Extraction honors .gitignore and .synapticignore, skips common secret files, and stays offline unless you explicitly enable a network-backed feature.

Connect your coding assistant

sh
synaptic install claude                    # Claude Code
synaptic install codex                     # Codex CLI
synaptic install codex --global            # Codex desktop app

Claude and Codex load only a small set of tools at startup, then find the rest on demand. This keeps prompts smaller without hiding capabilities. Gemini, Cursor, Copilot, OpenCode, Kilo, and generic AGENTS.md clients are supported too. See Assistant Integration.

What it does

NeedWhat Synaptic provides
Understand the codeSymbols, calls, imports, inheritance, resources, SQL, and cross-language boundaries in one graph
Find impactCallers, references, reverse dependencies, dynamic-dispatch hazards, and affected tests
Change code safelyChange forecasts, refactor plans, architecture diffs, and optional verification in an isolated worktree
Remember the pastSource-linked history from commits, decisions, incidents, reviews, CI, and previous attempts
Maintain dependenciesAPI contract tracking and vulnerability evidence with small, reviewable repair workflows

Synaptic supports dozens of languages, incremental updates, multi-repository federation, structural search, SQL auditing, graph-aware PR review, and exports for GraphML, Cypher, Graphviz, Obsidian, and Markdown. Detailed capabilities live in the documentation, keeping this page focused on getting started.

See the architecture

sh
synaptic chart

This creates a self-contained architecture map from the graph. Open a subsystem, select a symbol, and follow its real incoming and outgoing relationships without a server.

Built to save context

Synaptic returns the relevant slice of a graph instead of loading whole source files into an assistant's context. In the checked-in token benchmark, full graph answers used 27-38x fewer tokens than reading the source files referenced by those answers. That is a context-compression measurement, not a promise about total task cost. The methodology and raw results are in BENCHMARKS.md.

Common commands

CommandPurpose
synaptic extract .Build the graph
synaptic update / synaptic watchKeep it current
synaptic query "..."Find the relevant subgraph for a question
synaptic affected <symbol>Trace reverse impact
synaptic predict <files>Forecast risk and select tests
synaptic speculate <files>Verify a change in a throwaway worktree
synaptic memory search "..."Find relevant repository history
synaptic serveRun the MCP server

Run synaptic <command> --help for flags or use the complete command reference.

Pick a workflow

  • CLI: the fastest path for local extraction, queries, automation, and MCP.
  • Desktop: run synaptic-ui for visual repository setup, federation, assistant connection, updates, and the complete command catalog.
  • Hosted: Synaptic Cloud provides a managed MCP service. See the GitHub automation guide for commit-triggered graph sync and verified repair workflows.

Update a release installation with synaptic self-update.

Documentation

Build from source

The repository pins Rust 1.97.1.

sh
cargo install --path bin/synaptic
cargo install --path bin/synaptic-ui       # optional desktop app

For development and architecture notes, see the development guide.

Community and license

Questions, ideas, or something you built? Join the Discord community.

Synaptic is licensed under AGPL-3.0-or-later; see LICENSE and NOTICE. If you modify Synaptic and make it available over a network, you must offer those users the corresponding source. The separately maintained Synaptic Cloud service is proprietary and is not covered by this repository's license.

Use CodeGraph MCP with multiple AI models

TypingMind connects MCP tools at the workspace level, so once CodeGraph is connected, you can use it with different AI models in TypingMind instead of setting it up separately for each model. This MCP runs locally through the TypingMind MCP connector on your device.

Setup guide to use the local connector

Use this when the MCP server needs access to local files, apps, or private resources on your computer.

1

Open the MCP settings

In TypingMind, go to Settings, Advanced Settings, then Model Context Protocol and choose Setup Connector.

  1. Open TypingMind in your browser.
  2. Click the Settings icon.
  3. Go to Advanced Settings.
  4. Open the Model Context Protocol section.
  5. Click Setup Connector and choose This Device.
TypingMind MCP connector setup screen with This Device selected
2

Run the connector command

Choose This Device, copy the command from TypingMind, and run it in Terminal. Keep the process running while you use MCP.

  1. Copy the setup command shown by TypingMind.
  2. Open Terminal on macOS or Windows Terminal on Windows.
  3. Paste and run the command.
  4. Approve the package install if Terminal asks you to proceed.
  5. Keep the Terminal window running while using MCP tools.
3

Add CodeGraph as a server

When the connector status is Ready, click Edit Servers and paste the MCP server configuration.

  1. Wait until the connector status shows Ready.
  2. Click Edit Servers.
  3. Paste the CodeGraph MCP server configuration.
  4. Save the server list.
  5. Refresh if you want to confirm the connector is still ready.
TypingMind MCP settings showing active server and Edit Servers button
{
  "mcpServers": {
    "synaptic": {
      "command": "npx",
      "args": [
        "-y",
        "<mcp-server-package>"
      ]
    }
  }
}
4

Use it across models

Save the server list, open Plugins, enable the CodeGraph MCP tools, then select any supported AI model in TypingMind and use the tools in chat or assign them to an AI agent.

  1. Open the Plugins page in TypingMind.
  2. Enable the CodeGraph MCP tools.
  3. Start a chat and choose the AI model you want to use.
  4. Use the MCP tools in chat or assign them to an AI agent.
  5. Switch to another AI model whenever needed without reconnecting MCP.
TypingMind chat using enabled MCP tools with a selected AI model
Can you use CodeGraph to help me with this task?
CodeGraph
Sure. I read it.
Here is what I found using CodeGraph.

Frequently asked questions

What is the CodeGraph MCP server used for?

CodeGraph is an MCP server that lets compatible AI clients connect to external tools and context. In TypingMind, you can add this MCP server once and make its tools available in your AI workspace.

Can I use CodeGraph MCP with multiple AI models in TypingMind?

Yes. TypingMind connects MCP tools at the workspace level, so you can use CodeGraph with different AI models such as Claude, ChatGPT, Gemini, or other models you have configured in TypingMind without setting up the MCP server separately for each model.

Why use CodeGraph MCP with TypingMind?

TypingMind is one of the best frontends for LLM chat because it brings multiple AI models, prompts, plugins, AI agents, API keys, and MCP tools into one workspace. With CodeGraph connected, you can use its MCP tools across your preferred models while keeping your chat workflow organized in TypingMind.

How do I connect CodeGraph MCP to TypingMind?

CodeGraph runs through the TypingMind local MCP connector. This is best when the MCP server needs access to local files, desktop apps, command-line tools, or private resources on your computer.

What tools does CodeGraph MCP provide in TypingMind?

CodeGraph exposes MCP capabilities that can be enabled from the TypingMind Plugins page and used in chat or assigned to AI agents.

Do I need to share my API keys with TypingMind to use CodeGraph MCP?

No. TypingMind is local-first and lets you keep your model providers, API keys, prompts, and MCP configuration under your control. If CodeGraph requires authentication, add the required headers, OAuth settings, or local configuration for that MCP server when you create the connection.

Related MCP Servers

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇