Project Guardian

Project Guardian MCP is a focused memory and knowledge-graph powered server that helps you manage project data, relationships, and progress. It provides streamlined database operations alongside an intelligent memory system to keep your project context persistent, searchable, and actionable.

How to use

You interact with the Project Guardian MCP server through an MCP client. Use the memory and database tools to create entities (projects, tasks, people, resources), map relationships, and attach observations. Query, update, and export data as needed. Follow guided prompts to initialize memory, build your project graph, and perform routine health checks or progress updates. Use the memory system to store key observations and the graph to navigate dependencies, ownership, and blockers. The guidance system provides structured prompts to plan sprints, track progress, perform code reviews, and manage risks.

How to install

Prerequisites: Node.js version 18.0.0 or newer, npm, and SQLite3 (as a dependency). Install steps install dependencies, build, and then start the server.

# 1. Clone the repository
git clone https://github.com/1999AZZAR/project-guardian-mcp-server.git
cd project-guardian-mcp-server

# 2. Install dependencies
npm install

# 3. Build the project
npm run build

# 4. Start the server
npm start

Configuration snippets for client setup

{
  "mcpServers": {
    "project-guardian": {
      "command": "node",
      "args": ["/path/to/project-guardian-mcp-server/dist/index.js"],
      "env": {}
    }
  }
}

Additional configuration for MCP clients

{
  "mcpServers": {
    "project-guardian": {
      "command": "node",
      "args": ["/path/to/project-guardian-mcp-server/dist/index.js"],
      "env": {}
    }
  }
}

Available tools

execute_sql

Execute raw SQL queries on the in-memory SQLite database memory.db.

query_data

Query table data with filtering, pagination, and sorting.

insert_data

Insert records into a specific table in memory.db.

update_data

Update records in a table using defined conditions.

delete_data

Delete records from memory.db based on conditions.

import_data

Import data from CSV or JSON files into a table.

export_data

Export table data to CSV or JSON with optional filters.

initialize_memory

Initialize memory system and schema for the project knowledge graph.

create_entity

Create project entities (projects, tasks, people, resources) with metadata.

create_relation

Create relationships between entities (depends_on, owns, blocks, etc.).

add_observation

Add observations or notes to entities to capture progress and context.

delete_entity

Delete entities and their related relationships from memory.

delete_observation

Remove specific observations from entities.

delete_relation

Delete relationships between entities.

read_graph

Retrieve the full project knowledge graph with entities and relations.

search_nodes

Search for entities and relations by name, type, or content.

open_node

Get detailed information about one or more entities by name.