# OpenScaffold

> The Deployment OS: every customer deployment, and the coding agent building it, in one place. Free to sign up.

OpenScaffold reads everything a customer says (calls, emails, chat, tickets, docs), turns it into a plan where every item
quotes the customer, briefs your coding agent on it, checks the work against the customer's rules, and carries what
worked into the next deployment.

## For AI agents

- **A person asked you to set OpenScaffold up:** follow https://docs.openscaffold.ai/agent-quickstart.md from the top. It has
  every step. In a chat with no shell, tell them to add the connector `https://app.openscaffold.ai/api/mcp/mcp`.
- **A person wants to try it:** they sign up free at https://app.openscaffold.ai/sign-up (Google, GitHub or email; no
  credit card). Never create the account or click Approve for them.
- **Full documentation:** https://docs.openscaffold.ai/llms.txt

## What it does

- [Discovery Agent](https://openscaffold.ai/agents/discovery): Every call, email and ticket, read as it arrives and cited word for word.
- [Scoping Agent](https://openscaffold.ai/agents/scoping): A plan from the customer's own words, every line cited, ready for your sign-off.
- [Build Agent](https://openscaffold.ai/agents/build): Claude Code, Cursor or Codex, briefed on the customer before the first line.
- [Review Agent](https://openscaffold.ai/agents/review): Every change scored against what they asked for and the rules they set.
- [Playbooks](https://openscaffold.ai/agents/playbooks): What worked on one deployment, loaded into the next one of its kind.

## Who it's for

- [Forward Deployed Engineers](https://openscaffold.ai/for/forward-deployed-engineers): Every coding session starts briefed on the customer and held to their rules.
- [Solutions Engineers](https://openscaffold.ai/for/solutions-engineers): Calls, emails and tickets become a cited model of what the customer needs.
- [Implementation Engineers](https://openscaffold.ai/for/implementation-engineers): Each rollout starts with the playbook the last one left behind.
- [Deployment Strategists](https://openscaffold.ai/for/deployment-strategists): What the customer agreed to, traced to who said it, with changes flagged.
- [Deployment Leads](https://openscaffold.ai/for/deployment-leads): Where every deployment stands and what needs a decision, in one place.

## Pricing

Free to sign up, $0. Included:

- Every customer deployment in one place
- Connect calls, email, chat, support and docs, or upload files
- Your coding agent briefed on the customer: Claude Code, Cursor, Codex or any MCP agent
- Work checked against the customer's rules
- Playbooks that carry what worked to the next deployment

A bigger team, SSO or a contract: founders@openscaffold.ai. Details: https://openscaffold.ai/pricing

## FAQ

### How much does it cost?

It's free to sign up, and there's no credit card to enter. Add a deployment and see it work on your own customer.

### How is this different from a planning or context tool?

They stop at the plan. OpenScaffold keeps going: it briefs your coding agent on the customer, checks the work against their rules, and carries what worked into the next deployment.

### What happens when the customer changes their mind?

The plan changes with them. OpenScaffold catches it in the call or email, updates the plan from their words, and your agent's next session builds from the new version.

### How are playbooks different from a wiki?

Nobody writes them from scratch. As your agents finish work, they propose what worked; you approve it, and the next deployment of that kind starts with it loaded.

### Which coding agents does it work with?

Claude Code, Cursor, Codex or any MCP agent. Sign in once and your agent works from the customer's briefing.

### How do I get started?

Add a deployment and connect wherever the customer talks: calls (Granola), email (Gmail, Outlook), chat (Slack, Teams), support (Zendesk, Intercom, Pylon), and docs and tickets (Drive, Notion, Confluence, Linear, Jira, GitHub). Or upload files and call transcripts.

### Who is it for?

Anyone who takes customers live: forward deployed, solutions and implementation engineers, deployment strategists and leads.

## Blog

- [Better customer context, not more of it](https://openscaffold.ai/blog/retrieval-benchmarks): On 28 customer tasks, OpenScaffold got 18 points more answers fully right than standard RAG with the same context budget, and beat RAG given four times as much.

## Pages

- [Home](https://openscaffold.ai/) · [Pricing](https://openscaffold.ai/pricing) · [Docs](https://docs.openscaffold.ai) · [Support](https://openscaffold.ai/support) · [Privacy](https://openscaffold.ai/privacy) · [Terms](https://openscaffold.ai/terms)