MCP Catalogs
Homedreamgraph screenshot

dreamgraph

by mmethodz·82·Score 46

DreamGraph is a graph-first cognitive layer that builds a persistent knowledge graph to reason about code architecture.

knowledge-graphdeveloper-toolsai-llm
23
Forks
1
Open issues
this month
Last commit
2d ago
Indexed

Overview

DreamGraph is a comprehensive MCP server that combines a daemon, CLI, VS Code extension, and dashboard with a persistent knowledge graph. It maintains structured relationships between features, workflows, data models, and architecture decisions, allowing the system to answer from accumulated project understanding rather than just single file reads. It supports multi-repo systems, database introspection, and plugin development through its SDK.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Repository architecture understanding across multi-repo systems
you:Cognitive reasoning for code changes and documentation generation
you:Persistent knowledge maintenance in AI-assisted development
you:What programming languages does DreamGraph support?
you:How does DreamGraph maintain knowledge persistence?

When to choose this

Choose DreamGraph if you need to maintain architectural memory across multiple repositories and want a persistent knowledge graph for reasoning about code changes and system architecture.

When NOT to choose this

Avoid DreamGraph if you need lightweight integration with existing tools, have limited resources to run additional services, or prefer a simpler solution without persistent graph storage.

Tools this server exposes

12 tools extracted from the README
  • scan

    Perform a full scan of the repository to build the knowledge graph

  • enrich

    Enrich the existing knowledge graph with additional information

  • curate

    Curate the knowledge graph to improve quality and accuracy

  • query

    Query the knowledge graph for specific information

  • bootstrap

    Bootstrap the knowledge graph with initial data and structure

  • scan_database

    Scan and analyze the database schema to enrich the knowledge graph

  • status

    Check the status of the DreamGraph instance and knowledge graph

  • attach

    Attach a repository or workspace to a DreamGraph instance

  • start

    Start the DreamGraph daemon with specified transport

  • dream

    Execute a dream cycle to reason about and process the knowledge graph

  • export

    Export the knowledge graph or parts of it for external use

  • fork

    Create a fork of the current DreamGraph instance

Note: Tool names inferred from CLI commands and documentation. The README references an MCP tool surface but doesn't provide explicit tool documentation in this excerpt. Tools are primarily derived from CLI commands described in the Quick Start a

Comparable tools

architect-mcpsemantic-similarrefactcodebase-mcp

Installation

Install from source

**Windows (PowerShell):**

git clone https://github.com/mmethodz/dreamgraph.git
cd dreamgraph
./scripts/install.ps1 -Force

**Linux / macOS (Bash):**

git clone https://github.com/mmethodz/dreamgraph.git
cd dreamgraph
bash scripts/install.sh --force

For Claude Desktop integration, add to your claude_desktop_config.json:

{
  "mcpServers": {
    "dreamgraph": {
      "command": "dg",
      "args": ["mcp"]
    }
  }
}

FAQ

What programming languages does DreamGraph support?
DreamGraph is built with TypeScript and provides MCP tools that can analyze code in any language. It can work across repositories with different programming languages and technologies.
How does DreamGraph maintain knowledge persistence?
DreamGraph builds and maintains a persistent knowledge graph that records relationships between features, workflows, data models, architecture decisions, and other project elements. This graph survives individual sessions and provides long-term memory.

On Hacker News

Recent discussion from the developer community.

Compare dreamgraph with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.