# Space Setup Guide Step-by-step recipes for setting up common Space configurations — SEO workspace, development environment, content hub, and more. # Space Setup Guide This guide provides step-by-step recipes for setting up Spaces for common use cases. For general Space concepts, see [Spaces & Workspaces](https://docs.teamday.ai/guides/spaces). --- ## SEO Workspace A workspace for an SEO analyst with Ahrefs, Google Analytics, and Search Console access. ### 1. Create the Space ```bash teamday spaces create --name "SEO Office" ``` ### 2. Add an SEO Agent ```bash teamday agents create \ --name "Sarah" \ --role "SEO Analyst" \ --category marketing \ --system-prompt "You are Sarah, an SEO specialist. Analyze search performance, find keyword opportunities, and provide actionable recommendations. Always cite data sources and use tables for comparisons." \ --visibility organization ``` ```bash teamday spaces add-agent ``` ### 3. Install MCP Integrations If your organization has Ahrefs and Google Analytics MCP instances: ```bash teamday spaces add-mcp ``` ### 4. Store API Keys ```bash teamday spaces set-secret AHREFS_API_KEY=your-key GA_PROPERTY_ID=12345678 ``` ### 5. Add CLAUDE.md Create a `CLAUDE.md` in the Space root with project context: ```markdown # SEO Office Primary domain: example.com Focus keywords: [your target keywords] Competitors: competitor1.com, competitor2.com ## Reporting Schedule - Weekly: keyword ranking changes, traffic trends - Monthly: full SEO audit, competitor analysis ## Tools Available - Ahrefs: backlinks, keywords, domain rating - Google Analytics: traffic, user behavior - Google Search Console: impressions, clicks, CTR ``` --- ## Development Environment A workspace for code review, bug fixes, and development tasks. ### 1. Create from Git ```bash teamday spaces create \ --name "Backend API" \ --type git \ --git-url https://github.com/your-org/your-repo.git ``` ### 2. Add a Code Review Agent ```bash teamday agents create \ --name "Code Reviewer" \ --role "Senior Software Engineer" \ --system-prompt "You are a senior engineer. Review code for quality, security, and maintainability. Check for OWASP top 10 vulnerabilities. Provide specific fixes with code examples." ``` ```bash teamday spaces add-agent ``` ### 3. Add Skills ```bash teamday spaces add-skill core:research-assistant core:code-reviewer ``` ### 4. Store Credentials ```bash teamday spaces set-secret \ GITHUB_TOKEN=ghp-abc123 \ DATABASE_URL=postgres://user:pass@host/db ``` ### 5. Add CLAUDE.md ```markdown # Backend API Tech stack: Node.js, Express, PostgreSQL, Prisma ORM Deployment: Docker on AWS ECS CI: GitHub Actions ## Coding Standards - TypeScript strict mode - Tests required for all new endpoints - Use Prisma for all database operations - Follow REST conventions (proper status codes, resource naming) ## Important Paths - src/routes/ — API routes - src/models/ — Prisma schema and models - src/middleware/ — Auth, validation, error handling - tests/ — Jest test files ``` --- ## Content Creation Hub A workspace for blog writing, image generation, and content management. ### 1. Create the Space ```bash teamday spaces create --name "Content Hub" ``` ### 2. Create a Content Writer ```bash teamday agents create \ --name "Alex" \ --role "Content Writer" \ --category marketing \ --system-prompt "You are Alex, a content writer for B2B SaaS. Write in a conversational but professional tone. Use short paragraphs, clear headers, and data-backed claims. Always generate a cover image for blog posts." ``` ```bash teamday spaces add-agent ``` ### 3. Add CLAUDE.md ```markdown # Content Hub Brand voice: Professional but friendly. We explain complex topics simply. Audience: Marketing managers and business owners. ## Content Types - Blog posts (1500-2500 words) - Case studies - Product updates ## SEO Rules - Primary keyword in title and first paragraph - Meta description under 160 characters - At least 3 internal links per post ## Image Generation - Use landscape_16_9 for blog covers - Style: clean, modern, abstract - Save to images/ directory ``` The `MediaGeneration` platform tool is available automatically — the agent can generate images and videos without additional MCP setup. --- ## Multi-Agent Workspace A workspace with multiple specialized agents that collaborate. ### 1. Create the Space ```bash teamday spaces create --name "Marketing Team" ``` ### 2. Create Specialized Agents ```bash # SEO Specialist teamday agents create \ --name "Sarah" \ --role "SEO Analyst" \ --system-prompt "You are Sarah, an SEO specialist. When asked about SEO, analyze data from Ahrefs. For non-SEO questions, suggest handing off to the appropriate team member." # Content Writer teamday agents create \ --name "Alex" \ --role "Content Writer" \ --system-prompt "You are Alex, a content writer. Write blog posts and marketing copy. For SEO analysis, suggest handing off to Sarah." # Data Analyst teamday agents create \ --name "Max" \ --role "Data Analyst" \ --system-prompt "You are Max, a data analyst. Query data, generate insights, and build reports. For content creation, suggest handing off to Alex." ``` ### 3. Add All Agents ```bash teamday spaces add-agent ``` ### 4. Add CLAUDE.md with Handoff Instructions ```markdown # Marketing Team ## Team Members - Sarah: SEO analysis, keyword research, competitor analysis - Alex: Blog posts, marketing copy, social media content - Max: Data analysis, reporting, Google Analytics ## Handoff Protocol When a request is outside your expertise: 1. Use the UICommand tool to call listAgents 2. Find the right specialist 3. Use switchAgent to hand off the conversation 4. Briefly explain to the user why you're handing off ## Shared Resources - reports/ — Generated reports - content/ — Blog posts and articles - data/ — Analysis outputs ``` --- ## Automation with the CLI Script your Space setup for repeatable environments: ```bash #!/bin/bash # setup-workspace.sh SPACE_NAME="$1" AGENT_ID="$2" # Create space result=$(teamday spaces create --name "$SPACE_NAME" --format json) space_id=$(echo "$result" | jq -r '.space.id // .id') echo "Created space: $space_id" # Add agent teamday spaces add-agent "$space_id" "$AGENT_ID" # Add skills teamday spaces add-skill "$space_id" core:research-assistant core:data-analyst # Set secrets from environment teamday spaces set-secret "$space_id" \ OPENAI_API_KEY="$OPENAI_API_KEY" \ GITHUB_TOKEN="$GITHUB_TOKEN" echo "Space $space_id ready!" ``` ```bash chmod +x setup-workspace.sh ./setup-workspace.sh "My Project" "agent-abc123" ``` --- ## Next Steps - [Spaces & Workspaces](https://docs.teamday.ai/guides/spaces) — Core concepts and API reference - [Skills](https://docs.teamday.ai/guides/skills) — Build custom skills - [MCP Servers](https://docs.teamday.ai/guides/mcp-servers/what-are-mcp-servers) — Add tool integrations - [Missions](https://docs.teamday.ai/guides/missions-user-guide) — Schedule automated tasks