Server facts
ATLAS Task Management System MCP Server
Project
- Author
- cyanheads
- Repository
- github.com/cyanheads/atlas-mcp-server
- License
- Apache-2.0
- Category
- Productivity
Tech
- Runtime / language
- nodeTypeScript
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 477
- Last commit
- Data as of
Tags
- Knowledge Management
- Neo4j
- Project Planning
- Task Management
- Typescript
Overview
ATLAS is a project, task, and knowledge management server designed for LLM agents. Its model separates projects from their actionable tasks and supporting knowledge items. Projects record goals, statuses, links, completion requirements, and output expectations. Tasks add priorities, assignments, tags, and lifecycle tracking, while knowledge records hold contextual text, domains, and citations. Relationships and dependencies connect this material so agents can organize work around prerequisites rather than isolated checklist entries. Unified search spans the different entity types with property filters, fuzzy matching, and wildcard options.
Current versions use Neo4j, available through a self-hosted Docker setup or the AuraDB cloud service. The README identifies version 1.5.4 as the last SQLite release and describes the later three-node model as a replacement for the previous structure. Source installation requires dependency installation, a reachable Neo4j instance, environment configuration, and a build. MCP clients can then launch the compiled server over stdio, or use streamable HTTP with configurable host and port. The visible tool table explicitly documents creating, listing, updating, and deleting projects, including bulk operations. A basic web interface is marked experimental. HTTP configuration includes origin controls, authentication settings, and rate limits, but the README warns that production environment use has not been tested. That caveat matters when choosing between local agent integration and a remotely exposed deployment.
Install & configure
Requirements
- Node.js and npm
- Cloned atlas-mcp-server repository with npm install completed
- Build generated with npm run build
- Reachable Neo4j instance, self-hosted or AuraDB
node /full/path/to/atlas-mcp-server/dist/index.jsReplace the absolute entrypoint path after building. The README also provides npm run start:stdio and npm run start:http. HTTP defaults to 127.0.0.1:3010 with a documented /mcp endpoint. Production use is explicitly untested. Only project tool names are visible in the supplied excerpt.
Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.
Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Use Settings → Developer → Edit Config when available. Do not assume a Linux configuration path; check the client documentation for your platform. Remote servers use the app’s connector flow when supported, not the local stdio file shown here. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"atlas-mcp-server-cyanheads": {
"command": "node",
"args": [
"/full/path/to/atlas-mcp-server/dist/index.js"
],
"env": {
"NEO4J_URI": "<Neo4j connection URI>",
"NEO4J_USER": "<Neo4j username>",
"NEO4J_PASSWORD": "<Neo4j password>",
"MCP_LOG_LEVEL": "<Optional logging level>",
"MCP_TRANSPORT_TYPE": "<stdio for this local client configuration>"
}
}
}
}Config file · Project: .cursor/mcp.json
User: ~/.cursor/mcp.json
The project path is relative to the workspace root. The tilde denotes your user home directory. Use Cursor’s MCP settings to confirm which configuration and scope your installed version is reading. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"atlas-mcp-server-cyanheads": {
"command": "node",
"args": [
"/full/path/to/atlas-mcp-server/dist/index.js"
],
"env": {
"NEO4J_URI": "<Neo4j connection URI>",
"NEO4J_USER": "<Neo4j username>",
"NEO4J_PASSWORD": "<Neo4j password>",
"MCP_LOG_LEVEL": "<Optional logging level>",
"MCP_TRANSPORT_TYPE": "<stdio for this local client configuration>"
}
}
}
}Config file · Workspace: .vscode/mcp.json
For user scope, use the “MCP: Open User Configuration” command if available. Do not guess a user configuration path across operating systems, profiles, containers, or remote workspaces; check the client documentation. Merge this example with existing settings; do not overwrite unrelated servers.
{
"servers": {
"atlas-mcp-server-cyanheads": {
"type": "stdio",
"command": "node",
"args": [
"/full/path/to/atlas-mcp-server/dist/index.js"
],
"env": {
"NEO4J_URI": "<Neo4j connection URI>",
"NEO4J_USER": "<Neo4j username>",
"NEO4J_PASSWORD": "<Neo4j password>",
"MCP_LOG_LEVEL": "<Optional logging level>",
"MCP_TRANSPORT_TYPE": "<stdio for this local client configuration>"
}
}
}
}Config file · macOS/Linux: ~/.codeium/windsurf/mcp_config.json
Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json
Open the MCP configuration through Windsurf’s settings when possible to confirm the active file. Paths can differ in managed or future releases; check the client documentation if this location is not present. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"atlas-mcp-server-cyanheads": {
"command": "node",
"args": [
"/full/path/to/atlas-mcp-server/dist/index.js"
],
"env": {
"NEO4J_URI": "<Neo4j connection URI>",
"NEO4J_USER": "<Neo4j username>",
"NEO4J_PASSWORD": "<Neo4j password>",
"MCP_LOG_LEVEL": "<Optional logging level>",
"MCP_TRANSPORT_TYPE": "<stdio for this local client configuration>"
}
}
}
}Config file · Project scope: .mcp.json at the repository root
The project file uses mcpServers. For user scope, use claude mcp add --scope user rather than assuming .mcp.json in your home directory is the user configuration. Check the client documentation for managed settings and storage details. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"atlas-mcp-server-cyanheads": {
"type": "stdio",
"command": "node",
"args": [
"/full/path/to/atlas-mcp-server/dist/index.js"
],
"env": {
"NEO4J_URI": "<Neo4j connection URI>",
"NEO4J_USER": "<Neo4j username>",
"NEO4J_PASSWORD": "<Neo4j password>",
"MCP_LOG_LEVEL": "<Optional logging level>",
"MCP_TRANSPORT_TYPE": "<stdio for this local client configuration>"
}
}
}
}Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.
Tools
Source: repository README. Tool descriptions are a documented snapshot, not a live capability check.
| Tool | What it does |
|---|---|
| atlas_project_create | Create one project or a batch with metadata and dependencies. |
| atlas_project_list | List projects or retrieve details with filters and pagination. |
| atlas_project_update | Update one project or multiple projects in a batch. |
| atlas_project_delete | Delete a project or a batch of projects. |
Names, permissions, and defaults can change. Check the current upstream documentation before granting access.
Use cases
- Plan agent work as projects with completion requirements.
- Track prioritized tasks and prerequisite relationships.
- Keep cited knowledge connected to project work.
- Search across projects, tasks, and knowledge.
- Create or update projects in bulk.
Strengths
- Connects planning and knowledge in a shared graph model.
- Supports task and project dependencies.
- Offers stdio and streamable HTTP operation.
- Project operations include bulk modes.
Limitations
- Current versions require a Neo4j deployment.
- The basic web interface is experimental.
- The README says production environment use has not been tested.
Frequently asked questions
Independent listing, not an affiliation or security review. Grant the least access needed and verify the upstream instructions before running code.
