Now in Alpha — v0.2.0

You bring the ambition.
We bring the squad.

DashSquad lets you create AI agents — each with its own role. A researcher, a writer, an analyst. They work on their own, around the clock, even when you're not watching. Just chat with them like you'd message a coworker.

Download for Macor download manually →
$ curl -fsSL dashsquad.ai/install.sh | sh
DashSquad illustration
BUILT FOR TRUST

Runs on your machine.
Stays on your machine.

Your agents execute in isolated sandboxes on your own hardware. No cloud dependencies, no data exfiltration, full control.

Sandboxed ExecutionEach agent runs in an isolated environment with strict permission boundaries.
100% Local DataConversations and session data stay on your computer. Nothing leaves without your explicit command.
Full Audit TrailEvery action logged in append-only session files. See exactly what your agents did and why.
Zero SurveillanceNo tracking, no ads, no data selling. Your conversations stay between you and your agents.
Secure Sandbox
Research AgentRunning
Writer AgentProcessing
Analyst AgentRunning
All systems nominal
GETTING STARTED

Up and running in minutes

Deploy autonomous AI agents from Mission Control — no terminal needed. Configure, launch, and manage your entire squad from one dashboard.

Point-and-click deployName it, pick a model, select tools — your agent is live in seconds
Multi-agent orchestrationRun multiple specialized agents in parallel, each handling distinct tasks
Always-on executionAgents run autonomously in the background — check in when you want
Mission Control
Deploy New Agent
Agent Name
Research Assistant
Model
Claude 3.5 Sonnet
Tools
Web SearchFile ReadCode Exec
FLEXIBLE AI

Your AI, your choice.

Bring your own API keys — connect Anthropic, OpenAI, or Google directly with no middleman and no markup, or reach hundreds more models through a single OpenRouter key. Your keys, your usage, your bill.

Anthropic

Recommended

Claude models — exceptional reasoning, nuanced writing, and reliable tool use for complex tasks.

Google Gemini

Gemini models — multimodal intelligence with massive context windows and deep Google integration.

OpenAI

GPT models — versatile language models with broad capabilities and a massive ecosystem of integrations.

OpenRouter

One key, hundreds of models — DeepSeek, Llama, Qwen, Grok, and more, with automatic failover across providers.

YOUR WORKFLOW

Deploy. Run. Chat.

1

Deploy from Mission Control

Configure and launch agents with the visual deploy wizard. Select models, set tools, and deploy in seconds.

Research AgentActive
Writer AgentProcessing
Analyst AgentActive
2

Agents work autonomously

Your AI agents run independently, handling tasks and making decisions — just like real teammates.

3

Chat via your favorite apps

Interact through WhatsApp, Telegram, Slack, or any messaging platform you already use.