Questions about
Scope Architect?

Mission control for building with AI. Scope a project into agent-ready tasks, hand it to your coding agent over MCP or a cloud run, and stay in command as the build moves: context stays, progress is visible, and every decision is surfaced in one workspace.

Anything you build: an app, a marketing site, an API, an internal tool. A project holds as many scopes as it needs, and each scope breaks down into modules, features, and tasks. Every paid plan includes unlimited projects.

The MCP server is hosted, so there's nothing to install or pin. Clients with OAuth (Claude Code, Cursor, VS Code, Windsurf) press Connect. Claude Code CLI, Codex, and Gemini CLI take a one-line snippet with your key filled in. From there the agent reads the live plan, pulls the next ready task, and writes status, comments, and memory back as it builds. Skills for Claude Code, Cursor, and Codex install with one command, and any agent that doesn't speak MCP can take a JSON or Markdown export.

Pick a task and a cloud agent builds it in a fresh, isolated VM with your plan, stack, decisions, and acceptance criteria in hand. It streams its log, commits, and opens a pull request. The workspace persists on a saved disk between runs. Builder includes one concurrent run, Pro includes three, and Team includes one per seat. Extra slots are $25 a month each. The model bill is yours: connect your Claude subscription or your own API key.

Yes. Queue a whole module and the tasks run in dependency order. Set a budget in tasks, dollars, or minutes and the queue stops when it's hit. Gate each task on your approval, or let it run to the cap. A failed run pauses the queue instead of burning the rest. In your editor, the Autopilot skill runs the same loop through MCP with the scope as its only state.

A reusable set of preferences: your tech stack, target agent, code style notes, estimate unit (hours, story points, agent sessions), and testing preferences. The engine reads it on every generation so specs land in a context your agent already understands.

Yes. Book a walkthrough and we'll scope a real project with you and kick off a cloud run live, so you see exactly what you get. The demos and examples across the site show the rest.

Scoping your repo is free: sign up, connect GitHub, and get an agent-ready plan for your real code, then hand it to Claude Code or Cursor over MCP. That's one project and a one-time credit balance, no card. Builder adds cloud runs, unlimited projects, and a monthly credit refill. Want to see it on your project with a human first? Book a walkthrough.

Credits meter the AI work Scope Architect does for you. A full scope generation is 50, a chat message is 1, and an AI apply or edit is 5. Premium models such as Claude and GPT bill at five times those rates. Builder includes 1,000 a month, Pro includes 5,000, and Team includes 1,000 per seat, pooled. Bringing your own keys removes the meter entirely.

Yes, on every plan. Connect your Anthropic, OpenAI, Google, xAI, or DeepSeek keys and pay providers directly at their published rates, with no markup and no credit meter. Keys are encrypted at rest. Cloud runs on your own key stay subject to fair-use run-time limits.

Upgrade, downgrade, or cancel any time from Billing in the app. Annual billing is two months free.

Your workspace stays readable until the end of the period you paid for, and you can export any scope as JSON or Markdown at any time from the handoff panel. After that the account drops to the free limits; nothing is deleted automatically. Email support@scopearchitect.com and we will delete the workspace and its data on request.

Your scope, tasks, comments, and project context live in our database. Code is read for scoping when you import a repo and during cloud runs, which happen on a VM created for that run and torn down when it finishes. Agents run on your own keys or plans. We never train models on your content. The details, including subprocessors, are on the security page.

That's the point of Team. Invite people with magic links and pre-assigned roles. Humans and agents pick tasks from the same board, every change is attributed to whoever made it, and a field locks while someone is editing it so nobody races. When an agent hits a decision it can't make, it asks on the task and pauses. You answer there and the run resumes.

Card through Stripe, monthly or annual, with a receipt and invoice PDF for every charge. Team is $49 a seat, two seat minimum, and you set the count on the pricing page. Annual is two months free. Need a purchase order or a single invoice for a larger team? Talk to us under More.

You set the seat count on the pricing page, and you change it later from Settings → Billing. Adding or removing a seat is prorated. Each seat holds a run slot and adds 1,000 credits to the pool.

That is the expected path. One person scopes a repo free, moves to Builder or Pro, then switches to Team when other people join. The workspace, including scope, comments, memory, and agent history, carries over. Team starts at two seats.

Not yet. Accounts are email and password today. SAML single sign-on is on the roadmap for larger teams. If it is a hard requirement for you, tell us on a walkthrough call so we can prioritize it.

Yes. The security page covers data handling, storage, encryption, subprocessors, and deletion in one place, and the privacy policy covers GDPR and CCPA. A data processing agreement is available on request at support@scopearchitect.com. We are not SOC 2 certified yet; we will say so plainly and share our timeline if a deal needs it.