Your Personal AI Agent, Running on Your Hardware

Describe what you need done in plain language. An AI agent picks up the task, works through it step by step, and delivers the result — all running on infrastructure you control. Your data never leaves your machine unless you choose to connect an external service.

🏠

Self-Hosted & Private

Runs on your Mac, Docker server, or Kubernetes cluster. Your tasks, credentials, and data stay under your control.

🔄

Multi-Step Task Execution

Each task runs in an isolated container with its own AI agent. The agent reasons, calls tools, and delivers structured results — you watch it work in real time.

🔌

Integrations & Tools

Email, web search, GitHub, Slack, Google Drive, browser automation, and more — connected through the Model Context Protocol.

🧠

Persistent AI Memory

Agents remember context across tasks using Hindsight. Knowledge accumulates over time, so your agent gets better the more you use it.

🤖

Bring Your Own Model

Works with any LLM provider — Anthropic, OpenAI, Google, or local models via Ollama. Choose the right model for each task.

🔗

AI-Native & Extensible

Errand's MCP server lets other AI tools create and manage tasks. Connect Claude Code, Cursor, or custom agents to use Errand as a task execution backend.

Install on your Mac in minutes, or deploy to Docker or Kubernetes.

Get Started How It Works Blog GitHub