Skip to main content
Glama

create_project

Initiate a new project to track AI session handoffs and manage workflow transitions, enabling organized task prioritization and progress monitoring.

Instructions

Create a new project for tracking AI session handoffs

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
nameYesProject name
descriptionYesProject description

Implementation Reference

  • Core handler function in ProjectManager class that creates a new project: generates sanitized ID from name, checks for existing projects, creates ProjectMetadata, appends to projects list, saves metadata, initializes empty project data file, and returns the new project metadata.
    async createProject(name: string, description: string): Promise<ProjectMetadata> { const projects = await this.loadMetadata(); const id = name.toLowerCase().replace(/[^a-z0-9]/g, '_'); if (projects.some(p => p.id === id)) { throw new ProjectError('Project with this name already exists', id); } const newProject: ProjectMetadata = { id, name, description, created: new Date().toISOString(), lastAccessed: new Date().toISOString() }; projects.push(newProject); await this.saveMetadata(projects); await this.saveProjectData(id, { nextSteps: [], workingSessions: [], handoffs: [] }); return newProject; }
  • Input schema definition for the create_project tool, requiring 'name' and 'description' as strings.
    inputSchema: { type: "object", properties: { name: { type: "string", description: "Project name" }, description: { type: "string", description: "Project description" } }, required: ["name", "description"] }
  • src/index.ts:299-310 (registration)
    Tool registration in the list_tools response, defining name, description, and input schema for create_project.
    { name: "create_project", description: "Create a new project for tracking AI session handoffs", inputSchema: { type: "object", properties: { name: { type: "string", description: "Project name" }, description: { type: "string", description: "Project description" } }, required: ["name", "description"] } },
  • src/index.ts:427-437 (registration)
    Tool dispatch/registration in the CallToolRequestSchema handler: extracts arguments, calls the createProject handler, and returns result as JSON text content.
    case "create_project": const project = await projectManager.createProject( args.name as string, args.description as string ); return { content: [{ type: "text", text: JSON.stringify(project, null, 2) }] };

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/davidorex/project-handoffs'

If you have feedback or need assistance with the MCP directory API, please join our Discord server