Introduction
Tulsk is an agentic workspace where your team and AI agents work side by side. You get project management — workspaces, projects, tasks, comments — combined with multi-agent AI clusters that can research, plan, delegate, and execute work on your behalf.
Instead of switching between a chat tool, a project board, and a research tool, everything lives in one place. EMA, your AI Project Manager, oversees a team of specialized agents running on managed OpenClaw clusters.
Each agent has its own persona, tools, and workspace — and they can communicate with each other.
Why Tulsk?
AI and work are disconnected
Most teams brainstorm in ChatGPT, manage tasks in a separate tool, and do research in yet another. Context gets lost between tools, and AI outputs never become trackable work.
Tulsk keeps everything in one place. What you discuss with EMA becomes a real project with real tasks that agents can act on — and EMA knows what every agent has done.
Multi-agent teams, not single chatbots
Running one AI agent is useful. Running a team of specialized agents that can delegate work to each other is transformative.
Tulsk lets you create clusters of up to 6 agents on a single managed VPS — a Research Agent, a Content Writer, a Data Analyst — each with its own personality and skills.
EMA as your Project Manager
EMA isn’t just a chatbot. It’s your AI PM that:
- Context Awareness: Knows what all agents have been working on (activity injected into every conversation)
- Delegation: Can delegate tasks to the right agent based on skills
- Traceability: Traces delegation chains when agents hand off work to each other
- Automation: Generates scheduled reports automatically (“Every Monday, summarize project progress”)
- Integration: Pushes agent results back into your chat conversation
Agent setup is too complex
Running your own AI agents means managing servers, Docker containers, API keys, monitoring, and security. Most product teams don’t have the bandwidth for this.
Tulsk provisions and manages agent infrastructure for you. Create a cluster, give each agent a persona and skills, and they’re ready to work — with heartbeat monitoring, health checks, and one-click reprovision.
Project tools weren’t built for AI teammates
Traditional tools like Jira, Linear, or Notion are designed for humans tracking human work. They have no concept of assigning a task to an AI, letting it execute autonomously, or coordinating multiple agents.
Tulsk treats agents as first-class team members with their own assignments, run history, schedules, budgets, and delegation audit trails.