? Frequently Asked Questions

Everything people ask before they build.

Answers to common questions about Mission Squad — the agentic AI platform for building, managing, and deploying cooperative AI agents.

# What is Mission Squad?
Mission Squad is an agentic AI platform for building, managing, and deploying cooperative AI agents. Agents connect to any major model provider, work with your private data, use tools through the Model Context Protocol (MCP), and automate complex multi-step tasks — with a full trace of every run. It is a way of operating, not a chatbot: the work you keep re-explaining runs on a schedule while you do the work only you can do.
# How do I build agents — do I need to write code?
No code required. You describe what you want to Mission Squad's configuration agent and it builds the whole thing for you: it first reads your stack (your connected MCP servers and the tools they expose), proposes a plan — each agent, the model it runs on, the tools it needs, and how they hand off to each other — and once you approve, it writes the prompts, creates the agents, and wires up the workflow or factory. Changed your mind? Tell it in plain English and it reconfigures in place. See the getting started guide.
# Which AI model providers does Mission Squad support?
Mission Squad is model-agnostic. Point an agent at OpenAI, Anthropic, Google Gemini, Grok, Groq, Mistral, Cohere, Microsoft Copilot, OpenRouter, or ElevenLabs — or run self-hosted and open models on hardware you own. You can swap the model behind any agent in seconds, with no rewrite and no SDK to rip out.
# What is MCP and which tools can agents use?
MCP (Model Context Protocol) is the open standard Mission Squad uses to give agents tools. Connect a server, authorize it, and your squad can read and act in the systems you already use — Google Workspace, Gmail, Notion, Asana, Dropbox, Cloudflare, Shopify, Monday.com, Calendly, Klaviyo, Amplitude, Supabase, Make.com, CoinGecko, Beehiiv, Cloudinary, DeepWiki, Context7, Neon, and more. Connections use OAuth 2.1 with DCR, PKCE, and audited permission scopes. Read the MCP guide.
# Can I bring my own MCP server?
Yes. Drop in any OAuth-compatible MCP server and it shows up alongside the built-in connectors — your agents can use its tools the same way they use everything else. Details are in the MCP documentation.
# What are workflows and factories?
They are Mission Squad's two composition patterns. A workflow fans out: several agents take on the same request in parallel, and every response feeds back into one main agent that produces a single answer. A factory chains: steps run in sequence, each step's output becoming the next step's input — and any step can be a single agent or an entire workflow nested inside. Many systems use both. See workflows and factories in the docs.
# Can agents use my private data?
Yes. Mission Squad supports retrieval-augmented generation (RAG): create vector stores, upload your files, and configure agents to retrieve relevant context from them during a run. Agents grounded in your private data answer from your documents instead of hallucinating around them.
# Can I self-host Mission Squad on-premises or in my own VPC?
Yes. Deploy Mission Squad in your own VPC or on-prem so your keys, your data, and your traffic never leave your control. "Your data" means your data — not a polite word for someone's training set. See the self-hosting guide.
# How do I see what my agents actually did?
Every run keeps a full trace: every agent that fired, every tool it called, the exact arguments it passed, the data that came back, and per-step token accounting — kept in run history alongside every prior run. When a run does something you didn't expect, you don't guess — you open the trace and read exactly what happened.
# How much does Mission Squad cost to start?
Mission Squad is free to start — no credit card and no sales call required. Connect one tool, describe one agent, and read the trace yourself. Create an account to get going.
# Can agents run on a schedule?
Yes. Workflows and agents can run on schedules, so recurring work happens before you sit down — with the full trace waiting for you, not a status bar. You will know when your agents ran, and when they don't, you will know why.
# How do I get started?
Register for free, connect a tool, and describe your first agent to the configuration agent. If you're not up and running in an afternoon, we got it wrong and want to hear about it — tell us. The documentation covers everything in depth.