Atlas MCP Server
Project, knowledge, and task management system for AI agents via MCP.
Data last scanned yesterday · Reviewed 3 weeks ago
Overview
Atlas provides a structured architecture for LLM agents to track projects, organize tasks, and maintain a knowledge base. It uses a three-node data model connecting projects to their associated tasks and knowledge entities to ensure contextual consistency across workflows. Built in TypeScript, the system tracks specific project attributes like completion requirements and output formats alongside task priority levels and metadata. Developers can utilize these nodes to programmatically manage complex project lifecycles directly through their MCP-compliant AI tools.
Our verdict
This server receives a high-risk security score of 28/100, primarily due to it being a community-maintained project that has not undergone editorial review. Furthermore, the repository has not seen a commit in 390 days, which significantly impacts its long-term support and maintenance. Given these factors, it is best suited for experimental use in non-critical environments rather than integrated into sensitive or professional infrastructure.
- Uses a structured three-node data architecture
- Supports Apache-2.0 open source licensing
- Requires only 25 direct dependencies
- No code commits in over a year
- Lacks external security and editorial review
- Community-maintained without official vendor oversight
Setup
{
"mcpServers": {
"atlas-mcp-server": {
"args": [
"-y",
"atlas-mcp-server"
],
"command": "npx"
}
}
}Paste into ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart Claude Desktop. Some servers require API keys or tokens as environment variables — check the project's README and add an "env" object if needed.
- No read-only mode — running this grants full read/write access.
- No built-in authentication — anyone with access to this config can invoke every tool it exposes.
Tools
| Tool | Description | Risk |
|---|---|---|
| atlas_project_create | Creates new projects (single/bulk). | medium |
| atlas_project_list | Lists projects (all/details). | low |
| atlas_project_update | Updates existing projects (single/bulk). | medium |
| atlas_project_delete | Deletes projects (single/bulk). | high |
| atlas_task_create | Creates new tasks (single/bulk). | medium |
| atlas_task_update | Updates existing tasks (single/bulk). | medium |
| atlas_task_delete | Deletes tasks (single/bulk). | high |
| atlas_task_list | Lists tasks for a specific project. | low |
| atlas_knowledge_add | Adds new knowledge items (single/bulk). | medium |
| atlas_knowledge_delete | Deletes knowledge items (single/bulk). | high |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| ChatGPT | ||||
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf |
Frequently asked questions
›What database does the current version of the Atlas MCP server use?
Starting from version 2.0, the server uses Neo4j as its database, replacing the previous SQLite implementation.
›What are the hosting requirements for the current version of the server?
You must use either a self-hosted instance via the provided Docker Compose file or the Neo4j AuraDB cloud service.
›Does the system support linking information across projects and tasks?
Yes, the system supports knowledge integration where projects and tasks can be enriched with related information, as well as dependency management for workflows.
›How is the project data structured in the current version?
The system uses a 3-node architecture consisting of Projects, Tasks, and Knowledge items introduced in version 2.5.0.
Badge
Maintain this server? Add the live badge to your README.