User Guide

Got an idea? This coach turns it into a full product design through conversation. Paste → log in → start: that's the flow.

Connect your tool

Pick the AI tool you use and paste the text below into its chat as is. The tool walks you through connecting and logging in. One login stays signed in while you keep using it.

Copy this and paste it into your tool's chatAdd the PurplePrint MCP to Claude Code. Server name: purpleprint, transport: http, URL: https://app.purpleprint-ai.com/mcp. Register it globally (user scope), and add alwaysLoad: true to mcpServers.purpleprint in ~/.claude.json (this prevents /purpleprint:* slash commands from going missing in recent Claude Code IDE). If editing the file directly, save as UTF-8 no BOM. Walk me through Authenticate in /mcp. ⚠️ **Even if the /purpleprint:* slash commands never show up, it still works. Once connected, just say "start PurplePrint" in plain language: it runs the pp_start tool.** The slash list is a convenience, not the entry point. Don't reinstall or get stuck on it. If auth is needed and no prompt appears, disable and re-enable purpleprint, then check again. If it already shows "connected", auth is usually done. Also, so it resumes next time, append a resume rule to ~/.claude/CLAUDE.md (global): skip if already present, don't overwrite existing content: "When resuming a project you designed with PurplePrint, call pp_start first. If PurplePrint MCP is not connected, do not just proceed: first tell the user you are running without PurplePrint, show the connection guide (https://www.purpleprint-ai.com/en/docs#connect), and connect it for them if they want (the app must be reopened afterwards for the tools to appear). Ignore this for unrelated work." Final step (required): when setup and login are done, tell me exactly this. "Now restart Claude Code. In a terminal, end this session and run claude again; in the VS Code or Cursor extension, reload the window; in the desktop app, quit it completely (including the tray) and reopen it. Closing the window isn't enough: quit completely and reopen it so the new connection is picked up. Once it's open again, say "start PurplePrint"." A new connection only takes effect after the app is reopened, so it's normal if you can't confirm it in this session. Don't reinstall.
Copy this and paste it into your tool's chatAdd the PurplePrint MCP to this editor. Server name: purpleprint, URL: https://app.purpleprint-ai.com/mcp. Register it as a remote/http MCP in this tool's settings: globally (user settings, e.g. ~/.cursor/mcp.json or VS Code user settings), not just one project. If an OAuth/authorize flow is needed, walk me through the browser login. ⚠️ **Even if the command list never shows up, it still works. Once connected, just say "start PurplePrint" in plain language: it runs the pp_start tool.** The list is a convenience, not the entry point: don't get stuck on it. If auth is needed and no prompt appears, disable and re-enable purpleprint. If it already shows "connected", auth is usually done. Also, so it resumes next time, append a resume rule to ~/.cursor/rules/purpleprint.mdc (frontmatter alwaysApply: true) for Cursor, or your user instruction file for VS Code Copilot (global): skip if already present, don't overwrite existing content: "When resuming a project you designed with PurplePrint, call pp_start first. If PurplePrint MCP is not connected, do not just proceed: first tell the user you are running without PurplePrint, show the connection guide (https://www.purpleprint-ai.com/en/docs#connect), and connect it for them if they want (the app must be reopened afterwards for the tools to appear). Ignore this for unrelated work." Final step (required): when setup and login are done, tell me exactly this. "Now quit the editor (Cursor or VS Code) completely and reopen it. Closing the window isn't enough: quit completely and reopen it so the new connection is picked up. Once it's open again, say "start PurplePrint" in a new chat." A new connection only takes effect after the app is reopened, so it's normal if you can't confirm it in this session. Don't reinstall.
Copy this and paste it into your tool's chatAdd the PurplePrint MCP to Codex. Server name: purpleprint, URL: https://app.purpleprint-ai.com/mcp. Register it as a streamable HTTP MCP in ~/.codex/config.toml (shared by Codex CLI/IDE) and complete the OAuth login. Tell me up front to finish the login window within 5 minutes (Codex only waits 5 minutes). If it's late and a "You're signed in" notice or a 127.0.0.1 connection error appears, start the login once more: I'm already signed in, so it finishes quickly. On Windows PowerShell, if codex.ps1 is blocked by the execution policy, run it as cmd /c codex mcp ... When done, verify with codex mcp list. It's fine if /purpleprint:start doesn't appear in the / menu: saying "start PurplePrint" uses the pp_start tool. Also, so it resumes next time, append a resume rule to ~/.codex/AGENTS.md (global): skip if already present, don't overwrite existing content: "When resuming a project you designed with PurplePrint, call pp_start first. If PurplePrint MCP is not connected, do not just proceed: first tell the user you are running without PurplePrint, show the connection guide (https://www.purpleprint-ai.com/en/docs#connect), and connect it for them if they want (the app must be reopened afterwards for the tools to appear). Ignore this for unrelated work." Final step (required): when setup and login are done, tell me exactly this. "Now restart Codex. In the CLI, end this session and run it again; in the VS Code extension, reload the window; in the ChatGPT desktop app, quit it completely (including the tray) and reopen it. Closing the window isn't enough: quit completely and reopen it so the new connection is picked up. Once it's open again, say "start PurplePrint" in a new thread." A new connection only takes effect after the app is reopened, so it's normal if you can't confirm it in this session. Don't reinstall.
Copy this and paste it into your tool's chatAdd the PurplePrint MCP to Antigravity as a global MCP. Server name: purpleprint, URL: https://app.purpleprint-ai.com/mcp. Register it via serverUrl in ~/.gemini/config/mcp_config.json, and check ~/.gemini/antigravity/mcp_config.json if needed. After registering, go to Settings > Customizations > Installed MCP Servers, click Refresh, and complete Authenticate. Also, so it resumes next time, append a resume rule to the global instruction file Antigravity reads at session start (e.g. ~/.gemini/GEMINI.md or AGENTS.md) (global): skip if already present, don't overwrite existing content: "When resuming a project you designed with PurplePrint, call pp_start first. If PurplePrint MCP is not connected, do not just proceed: first tell the user you are running without PurplePrint, show the connection guide (https://www.purpleprint-ai.com/en/docs#connect), and connect it for them if they want (the app must be reopened afterwards for the tools to appear). Ignore this for unrelated work." Final step (required): when setup and login are done, tell me exactly this. "Now quit Antigravity completely and reopen it. Closing the window isn't enough: quit completely and reopen it so the new connection is picked up. Once it's open again, say "start PurplePrint"." A new connection only takes effect after the app is reopened, so it's normal if you can't confirm it in this session. Don't reinstall.

Use ChatGPT and Claude in their desktop apps (Codex in the ChatGPT app · Claude Code in the Claude app). The browser versions can't save documents, so you can't continue there.

After pasting, reopen the app. A new connection only takes effect after you quit the app completely and reopen it (closing the window isn't enough). In a terminal, start a new session; in an editor extension, reload the window; for a desktop app, quit it including the tray. Then say “start PurplePrint”.

Press the login button and this screen appears: one tap with Google or GitHub:

You need to log in to continue.
We'll verify with your account just once.

Continue with Google
Continue with GitHub

Stuck or unsure? We'll do it with you. In a free onboarding session we look at your screen together and get you connected. To sort it out yourself, see Troubleshooting.

Supported tools

The connection is remote HTTP MCP + OAuth. No API keys or config secrets needed.

ToolStatusHow
Claude CodeSupportedURL + alwaysLoad: true + /mcp Authenticate
CursorSupportedurl in user settings + browser OAuth
VS CodeSupportedAdd the Claude Code or Codex extension and it connects through that. Direct registration in user settings is in the tab below (Copilot's own MCP is being validated)
CodexSupportedcodex mcp add/login + natural-language pp_*
Antigravity (Google)Supportedconfig serverUrl → Refresh → Authenticate → paste code → restart (details below)
Devin Desktop · Cascade (formerly Windsurf) · Devin · LovableComing soonurl + OAuth confirmed in docs · connection being validated (Devin needs org permission)
EmergentExploringurl-based + DCR OAuth (validating)

Client UI

Find this spot inside your tool

Claude Code

> /mcp

MCP servers

Usage: /mcp [reconnect|enable|disable [<server>|all]]

/mcp

purpleprint

Remote HTTP · OAuth required · alwaysLoad true

filesystem

Connected · local stdio
Running

github

Disabled for this workspace
Disabled
Cursor Settings

Settings / Tools & MCPs

MCP Servers

Ppurpleprint

Remote HTTPS · https://app.purpleprint-ai.com/mcp
OAuth required
Add MCP Server
Remote HTTPSCommand
Edit JSONSave
mcp.json
{
  "servers": {
    "purpleprint": {
      "url": "https://app.purpleprint-ai.com/mcp"
    }
  }
}

The MCP Server Definition 'purpleprint' wants you to authenticate to PurplePrint.

Cancel
Add Client Registration DetailsDynamic Client Registration not supported · Client ID · Client Secret
Codex · config.toml
# ~/.codex/config.toml
[mcp_servers.purpleprint]
url = "https://app.purpleprint-ai.com/mcp"

$ codex mcp add purpleprint --url https://app.purpleprint-ai.com/mcp

$ codex mcp login purpleprint

Opening browser for OAuth...

purpleprint · remote · authenticated

Antigravity · Settings

Installed MCP Servers

Add MCP +Refresh ⟳
purpleprint

› 34 tools enabled

I'll set it up manually (without an agent)

Claude Code

  1. Register globally from the terminal: claude mcp add --transport http -s user purpleprint https://app.purpleprint-ai.com/mcp
  2. On recent IDE, add "alwaysLoad": true to mcpServers.purpleprint in ~/.claude.json: prevents missing slash commands.
  3. Open /mcp → Authenticate on purpleprint → browser login
  4. Restart Claude Code: in a terminal, end the session and run claude again; in the extension, reload the window; for the desktop app, quit it completely (including the tray) and reopen it

Cursor · VS Code

  1. Register the server in user settings (~/.cursor/mcp.json or VS Code user settings):
{ "mcpServers": {
    "purpleprint": { "url": "https://app.purpleprint-ai.com/mcp" }
} }
  1. Browser OAuth pops up on connect → log in
  2. Quit the editor completely and reopen it (closing the window isn't enough) → say “start PurplePrint” in a new chat

Codex

  1. On Windows: cmd /c codex mcp add purpleprint --url https://app.purpleprint-ai.com/mcp → cmd /c codex mcp login purpleprint
  2. It's fine if slash (/purpleprint:*) doesn't appear: say “start PurplePrint” and it uses the pp_start tool.
  3. Finish the login window within 5 minutes: Codex only waits 5 minutes for it. If you're late, a “You're signed in” notice appears; just start the login in Codex once more: you're already signed in, so it finishes quickly.
  4. Restart Codex: in the CLI, end the session and run it again; in the extension, reload the window; for the ChatGPT desktop app, quit it completely (including the tray) and reopen it → say “start PurplePrint” in a new thread

Quick start: first session

  1. Connect: In Connect your tool, pick your tool's tab and paste the text into its chat.
  2. Log in: Once in the browser with Google/GitHub. (stays signed in while you use it)
  3. Resume rule: The connect prompt also plants a “resume next time” rule in your global settings, so after a reboot or a new session the coach picks up automatically: and if PurplePrint isn't attached, it tells you first.
  4. Start: /purpleprint:start (or natural language “start PurplePrint” → pp_start). When the welcome shows, reply “let's start” and Phase 0 begins.

Already have writing·notes·Notion? Paste them in too: it pulls the core and backfills what's missing.

Here's how the first conversation flows:

First session

It's a demo: the real thing runs on your idea, inside your tool. Stuck? Just say “again” or “simpler” and it fixes course.

What the coach does for you

So you can go deep without knowing design, the coach recommends first, keeps it in documents, and checks again. Each item notes where it works (web coach · AI tools) and when it shows up.

It recommends first

Instead of asking you to pick, it shows the options, its first pick, why, and why the rest wait. You just confirm. Decisions only you can make, such as pricing, refunds, what personal data to collect, or product direction, are asked separately.

Phases 0 to 9 · web coach · AI tools

Design documents build up at every phase

From the context document through the design document, plan, UX design, wireframes, design system, review log, marketing strategy, handoff and task list, each phase leaves a document. Only what you confirm is finalized, and a phase in progress is marked as a draft.

Web coach = Docs tab · AI tools = .md files in your workspace

It looks things up and keeps the evidence

The coach searches and reads the web for competitors, the people you're building for, marketing and launch channels, and material for the features you need, including GitHub and Reddit, and keeps the sources in your documents. Deeper research runs on the research engines in the command reference.

Phases 0 to 9 · web coach · AI tools · See the research engines →

A design map shows the core at a glance

Even as documents pile up, the value chain, user journey, screens, core logic and data relationships are drawn as a map straight from your documents.

Web coach

Pin it, fix it, talk it through

Click a spot on the map and leave a line. Pin several and send them at once; the coach goes to each spot and either fixes the document or talks through why it was decided that way.

Web coach

Recursive checks every session

It doesn't stop after one fix. It rechecks where documents disagree and repeats until two passes in a row find no new gaps, and the rounds stay in the review log. While you build with it connected to your AI tool, it also checks design documents, development design and code against each other each session. Two clean passes mean nothing new was found within what was checked, not a guarantee of zero defects.

Phase 6 · while building

It proposes the order of development design

After the service design, it proposes what to design and build first with options, a first pick and its reason, and why the rest wait. It also settles who builds it and what goes into the first scope.

Phases 8 to 9

A task list keeps progress and backlog

The task list sets what to build, in what order, and when each task counts as done. Once you start building, the coach keeps progress up to date and collects new to-dos in the backlog before promoting them to tasks.

After Phase 9 · while building

It proposes task priority first

Before you ask what to do next, it names the first priority with its reason, plus what to postpone and why. When you reopen a session, it starts from progress and the one next task.

After Phase 9 · while building

The harness handles session memory and context

No need to write or maintain rule files like CLAUDE.md or AGENTS.md. Each session the coach reads the one-page summary (project-lens) and the task list first, and when a conversation gets long it wraps up and hands over to a new session. The web coach keeps your conversations and documents in your account.

AI tools · web coach

It shows how you grow as a builder

Your 9-cell builder level updates each time you finish a design document. Leading stage, the 7 signals and the next step show how well you direct AI and what to practice next.

Web coach · AI tools · About builder levels →

How to read a Hyper-Review report

Read it in this order: identity verdict (can it be said in one sentence), design map (gaps across the ten parts), structure map (root issues and what they unlock), dig into evidence (every finding, the root cause chain, team capability), then actions and handoff (next actions, readiness gate, handoff messages).

Hyper-Review · See the sample report →

Phase 0~9 flow

This is the heart of PurplePrint: a 10-step coaching flow that turns a daunting one-line idea into a build-ready design. So you never start alone and stuck, the coach leads one step at a time, one question at a time. Each step stacks on the previous one's conclusion, building the “why→who→what→how” purpose chain with no gaps in the middle.

How it goes: see for yourself

Try the tabs. The coach closes the gap first, you answer in plain words, and if you don't like it you can push back·debate: each of these is a way to use the product. It's not pressing “sounds good” like a general AI: you lead.

You never start staring at a blank screen. At every step the coach proposes options with a top pick and reasons, first.

purpleprint · design session

And then, a whole session end to end

How those usage modes actually play out: I followed a whole session, from Phase 0 to 9, on the one line “a cover-letter service.” Scroll and each step plays, and when a step finishes the artifact appears right there.

purpleprint · design session
ArtifactPreview

Command reference

Below is the /purpleprint:<command> form for slash clients (Claude Code·Cursor). Tool clients (Codex) use the same features as pp_<command>: e.g. /purpleprint:status = pp_status.

Start · Progress

/purpleprint:startStart or resume: the first command you use. New? A warm onboarding; returning? The server remembers your last Phase·artifacts·next action and picks up right there.
/purpleprint:Step progression (0–9)You don't pick steps from a list: start, and the coach reads your state and leads you through context(0)·persona(1)·problem(2)·product(3)·UX(4)·wireframe(5)·check(6)·GTM(7)·dev design(8)·handoff(9). To jump to a specific step, just say “let's start from Phase 4”.

Status · Ops

/purpleprint:statusShows at a glance which Phase you've reached, what artifacts you've made, and your next action.
/purpleprint:listLists the artifacts you've made so far (context doc·design doc, etc.).
/purpleprint:help-next-todoStuck? It points to exactly one next action from where you are.
/purpleprint:projectsGot several ideas? See your project list to switch or create a new one.
/purpleprint:saveSaves a checkpoint here. Pick up exactly there next time with start.
/purpleprint:how-to-useFirst time? A quick how-to for PurplePrint.
/purpleprint:manualA more detailed install·auth·troubleshooting manual. Reach for it when server-health doesn't resolve it.
/purpleprint:server-health“Connected but why isn't it working?”: self-diagnoses server reachability·source version·auth status and tells you per-tool recovery.
/purpleprint:oauthBriefs your login status·tier·remaining trial period.
/purpleprint:handoffBundles your finished design artifacts into a package to hand a developer or AI directly.

Growth

/purpleprint:my-growthYour nine-cell builder level updates automatically each time you finish a design document: nothing to invoke. Call it yourself to sample your chat logs so far and see, day by day, how your builder capability has grown. It reads behavior signals only, not chat content (content-free).

Kits

/purpleprint:engine-listSee the Kits you can build on right away: templates for service design, development design, web pages, apps and marketing content. The coach may also suggest the Kit that fits your phase.
/purpleprint:engine-installDownload the chosen Kit into your project folder with its original file structure. Large Kits come in parts. Available on a free trial or subscription.

Research & validation engines

The PurplePrint design coach draws your design out from what you answer. But things like “how do competitors actually do it?” or “does this persona really exist?” are information that isn't in your head. The research & validation engines fill that blank with real investigation, so your design stands on evidence rather than guesses.

What's fundamentally different from general AI: General AI (ChatGPT·Claude, etc.) makes you ask for everything (“research the competitors”) and skips it if you don't. PurplePrint proposes, inside the Phase 0~9 framework, at the right moment: “now would be a good time for this research”. The coach flags what you're missing before you even know it's missing: that's the fundamental difference.

/purpleprint:hyper-researchDeeply researches competing products·tech·market → so you know “how others already solved this” and find a differentiator you won't just copy. (mainly Phase 2)
/purpleprint:hyper-research-marketingResearches competitors' channels·copy·content·community → so you start with evidence for which channels·messages land, instead of burning ad budget blindly. (mainly right after Phase 6)
/purpleprint:hyper-research-personaValidates with data whether your assumed persona actually exists → cuts the risk of betting on an “imaginary user,” and finds adjacent markets you missed. (mainly Phase 1)
/purpleprint:hyper-scenarioSimulates your design as the flow a real user goes through → catches missing scenarios·break points before you build, reducing rework. (mainly right after Phase 5)
/purpleprint:hyper-research-gtmResearches search demand·competitor channels·launch cases → so you start with a strategy for where and how to get your first users at launch. (mainly Phase 7)

How it's called: take the suggestion, call the command, and it runs on your AI agent's web capability (₩0 extra). By default the research itself is handed to a separate helper agent (Codex): when research material piles up in the coaching thread, the design decisions that follow get dragged along by it. If you have a Codex subscription, the coach walks you through install and login, then runs the helper for you and only reads back the result (no need to open the Codex app). If you don't have one, the coach just does the research itself. Or call it yourself anytime. The “mainly Phase N” on each line is the best-fit moment, not the only one.

/purpleprint:hyper-review high tier

① Service design: persona·value chain·revenue·UX·security·team capability across 39 axes, recursing until zero findings. Answers 'is it a real product' at expert depth on the server. (mainly Phase 6)

/purpleprint:hyper-review-dar high tier

② Dev design: architecture·contracts·data models reviewed across 8 rotating lenses (DAR). Answers 'will the structure hold' before you write code. Only design docs go to the server, never the code. (mainly Phase 8)

/purpleprint:hyper-review-code-audit local

③ Code: using the earlier Review·DAR reports as the baseline, checks whether the code matches the design. Code never goes to the server (your AI agent audits locally · zero cost).

The three tiers (service design → dev design → code) are one review ladder. Each tier can go deeper with Ultra: lifting the round cap and recursing until findings hit zero (not a smarter AI, just a more relentless dig).

Here's how a Hyper-Review series run progresses (dev-design DAR example):

purpleprint · Hyper-Review DAR

› pp_hyper_review_dar (dev-design docs as input)

# Hyper-Review DAR (dev-design architecture review) started

- jobId: dar-7f3a… · mode: dar (dev-design docs)

Check progress/result with hyper-review-status.

› pp_hyper_review_status (a while later)

# Hyper-Review running

- Progress: round 3/8 (rotating deep review): 7 findings so far

- Elapsed: ~6 min · usually about 30 minutes

› pp_hyper_review_status (on completion)

# Hyper-Review complete

- Findings: 17 → 5 core issues

## Root cause

identity → value chain → async experience gap → missing billing cap → weak observability

## Core issues

1. [high] Undefined state transition when the billing cap is exceeded

2. [high] Metering atomicity = observation grade (not billing grade)

3. [medium] Hardcoded model prices …

## Next actions · team handoff

The report is saved to your local (docs/review/): with teammate·AI handoff messages included.

A demo: in practice it runs on your design docs, across the three tiers (service-design Review · dev-design DAR · Code Audit) the same way. Only Code Audit keeps the code off the server: your AI agent audits it locally.

Shared links

A link anyone can use to view a web coach project without logging in. Create one with “Share” in the project menu and manage it under Workspace in the Settings tab.

The web coach is the design coach you use in the browser, with nothing to install.

What the link shows

  • This project's design documents (Documents tab) and progress (Progress tab).
  • Only for links with “Show the conversation too” on: your conversation with the coach, including past ones. Your messages, the coach's replies, the options you picked, one line per tool used, and results written to repos or Notion.

What it doesn't show

  • Account details such as your login email.
  • Usage numbers and cost.
  • Your connected GitHub and Notion accounts and their tokens.
  • Any other project.
  • The conversation, for links with that option off.

Who can see it

  • Anyone with the link can view it without logging in. Choose who you send it to.
  • Viewers can only read. They can't talk to the coach or edit documents.
  • It's blocked from search engines.

It shows the current state

  • The link doesn't freeze a copy. It shows the documents and progress as they are when it's opened.
  • Links that show the conversation also show conversations after the link was made.

Expiry and stopping

  • Choose 7 days, 30 days, or Never when you create it.
  • Stop it anytime under Workspace in the Settings tab. A stopped link closes right away.
  • Archiving or deleting the project closes its links too.

Privacy

Where your content lives depends on where you use the coach. First, four promises that hold wherever you use it.

Not used to train AI

AI processing for the web coach and Hyper-Review is not used to train models. External AI models are called only through zero data retention routes that keep neither input nor output. Privacy policy, Section 6

Stays on your computer with your AI tool

Connected to Claude Code, Cursor, and similar tools, coaching conversations and design documents stay in your workspace and never reach PurplePrint's servers. Privacy policy, Section 5

Your designs are yours

You own the artifacts you create with the coach, such as the context and design documents. Terms, Section 9

Delete means deleted

Deleting a web coach project removes its conversations and design documents right away and stops its share links. Privacy policy, Section 6F

Exactly what lives where, by where you use it:

In your AI tool: content-free

When you connect it to your own AI tool, like Claude Code, Cursor, or Codex. Our server doesn't see what you're building or what you wrote.

Local only: never to the server

Your idea·chat·design docs (context doc·design doc·UX doc, etc.) all live on your computer only. Research-engine results too: they run on your AI agent's web capability, so the content never passes through our server.

What goes to the server: progress meta only

Only progress signals like “which Phase you reached” or “how many times you used a command.” What the idea is, or what you answered: that content never goes. It's the minimum needed for resume·usage management.

Growth measurement (my-growth) is content-free too

Builder capability is measured from behavior signals (like which decisions you made yourself) without chat content. It doesn't look at what you said.

In the web coach: kept on our server

The web coach is the design coach you use in the browser, with nothing to install. With no computer of yours to hold the files, it keeps your documents and conversation on our server.

What's kept: documents and conversation

The design documents you make in the web coach and your conversation with the coach (including past ones) are kept in our server storage, so you can pick up right where you left off.

AI processing: zero-retention routes only

To write the coach's replies, the conversation goes to AI processing. Only routes that keep neither the input nor the output (zero data retention) are used, and it is not used to train models.

When other users can see it: only if you share

Other users can see it only when you create a shared link. The conversation shows only on links with “Show the conversation too” on.

When the Company looks: only in set cases

As a rule the Company does not open your content. Designated staff look only when you ask us to resolve something, when an outage or security issue needs it, or for quality checks (you can opt out in Settings), and every look is logged. Privacy policy, Section 6I

Deleting removes it right away

Delete a project and its documents and conversation are deleted right away, and its shared links close. Only a progress record with no content remains.

Trial · pricing FAQ

Do I need to know how to code?

The design work itself needs no coding. Answer in plain words even without design terms: at each step it proposes 2~3 options and a top pick first, so you never start staring at a blank screen. Confused? Just say “again” or “simpler” and it fixes course right away. That said, PurplePrint runs inside an AI coding tool (Cursor, Claude Code, etc.): it helps if you're already using one.

I already have notes·a Notion doc.

Paste them and it pulls the core, normalizes it, and backfills the missing “why·who·what problem.” You don't rewrite from scratch: it fills the gaps on top of what you have.

Where are the artifacts saved?

They stack as `.md` files in your local project folder (context·design·UX docs, etc.). Next time you `start`, the server remembers your last point and resumes right there. If you make them in the web coach, they're kept on our server and you see them in the web coach's Documents tab.

Which tool should I use it in?

Claude Code, Cursor, Codex, and Antigravity are supported. In VS Code, add the Claude Code or Codex extension and it connects through that. Devin Desktop·Cascade (formerly Windsurf), Devin, and Lovable are coming soon; Emergent is still being validated. If setup feels heavy, you can also run a review on the web. Detailed connection is in [Connect your tool] above, per tool.

Do the research engines cost anything?

Research·Scenario·Marketing·Persona·GTM run on your own AI agent, so PurplePrint adds ₩0. If the research is handed to a Codex helper agent, it runs on your Codex subscription (and if you don't have one, the coach does the research itself). Only Hyper-Review (39 axes), which uses a server LLM, is high tier.

Is it free?

Sign up and a 2-week free trial starts automatically, no card. After that, beta price $13/mo (list $20): an add-on next to your existing AI subscription.

Why do I need to log in?

It's the minimum for account identity and trial/subscription management. Log in once with Google or GitHub and it stays signed in while you keep using it (at least 90 days), so you don't repeat it. Sign in the same way you signed up (email sign-in is only for accounts issued in advance).

Do my ideas·code go to the server?

Not when you use it in your AI tool (Claude Code, Cursor, Codex, etc.). Your chat·artifacts stay on your machine only, and the server gets just progress meta like “which Phase you reached” (content-free). The web coach (the coach you use in the browser) keeps your documents and conversation on our server so you can pick up again. More in [Privacy] above.

Codex login ends with “127.0.0.1 refused to connect.”

If the login takes more than 5 minutes, Codex stops waiting and closes the connection. Your account is already created: just start the login in Codex once more; you're already signed in, so it finishes quickly.

I don't see slash (/) in Codex.

That's normal. Codex is a tool surface: just say “start PurplePrint” and it uses the `pp_start` tool. Stuck? Diagnose with `pp_server_health`.

Troubleshooting

I just installed it, but login isn't picked up or there are no commands

The session you installed from doesn't re-read the new connection or login. Quit the app completely and reopen it and it's picked up. No need to reinstall.

  • Desktop apps (Claude, ChatGPT, Cursor, Antigravity): quit completely, including the tray, then reopen (closing the window isn't enough)
  • Terminal (Claude Code, Codex CLI): end this session and run it again
  • Extensions in VS Code or Cursor: Command Palette → Reload Window

Once it's open again, say “start PurplePrint”. If it still doesn't work, see the items below.

The sign-in page shows “Forbidden” or asks for a password

Sign in the same way you signed up: Continue with Google if you used Google, Continue with GitHub if you used GitHub. Continue with email is only for accounts PurplePrint issued in advance (reviewers and operators), so typing a Google address there won't work.

Signing in again keeps failing the same way

The connection saved in your AI tool is out of date. Instead of retrying, remove the connection and add it again, and you'll pick up right where you left off. Your designs and account are unchanged.

  • Claude Code: claude mcp remove purpleprint → claude mcp add --transport http -s user purpleprint https://app.purpleprint-ai.com/mcp → fully restart the app, then Authenticate in /mcp
  • Other tools (Cursor, Codex and more): remove PurplePrint from the MCP settings, add it again, reopen the app, and sign in

After a reboot, it just proceeds without PurplePrint

After a reboot or in a new session, the MCP may not re-attach automatically (and your login can drop). Then it quietly runs as a plain AI: it looks the same, so it's hard to notice.

Prevent it: With the resume rule the connect prompt plants in your global settings, a new session has the coach call pp_start automatically, and if it isn't attached it tells you “running without PurplePrint” first and walks you through reconnecting. If you haven't planted it yet, paste the Connect prompt again, or tell the coach “plant the resume rule”.

Claude Code (advanced): For an extra guarantee, add a SessionStart hook to ~/.claude/settings.json. Every session start it runs claude mcp list to check the real connection state and tells you deterministically if it's not attached (rules rely on the AI's judgment; the hook detects actual state). Ask the coach to “plant the PurplePrint connection-check hook (in node)”.

I want to continue on another computer

Your design documents live in the project folder on the computer where you started. To continue with the same documents on another computer, keep them in a GitHub repository. When your first document is ready, the coach offers once to create one (you can just keep going if you don't want it).

  1. Connect PurplePrint in that computer's AI tool: paste the Connect prompt above and reopen the app.
  2. Ask to continue: say “Get [repository URL] and continue with PurplePrint”, and the AI tool pulls the documents and picks up where you left off.

Projects you started in the web app continue on any computer once you sign in.

Slash commands (/purpleprint:*) don't show

Connected (✓ Connected) but commands don't appear? In order:

  1. Claude Code: Open ~/.claude.json and check "alwaysLoad": true exists inside mcpServers.purpleprint. Add the line if missing. (Recent IDE may not surface slash in a fresh window without it.)
  2. Reload: Refresh the window, or type /mcp → Reconnect on purpleprint. The command list reloads.
  3. Still nothing: Fully restart the editor and open a new chat.

Disconnect is usually unnecessary: the server always keeps the list open, so reload/Reconnect is enough.

“Connected” but commands don't work (“auth required” / 401)

Commands show but execution is blocked = token expired or not attached (the server is fine). Just log in again in your tool:

  • Claude Code: /mcp → purpleprint → Authenticate (or Reconnect) → browser login
  • Cursor: Settings → Tools & MCP → purpleprint → Authenticate. If it looks connected, hit Logout once, then log back in.
  • VS Code: In MCP Servers, purpleprint → Login with PurplePrint
  • Codex: Open a new thread or restart, then “start PurplePrint”

One login stays signed in while you keep using it (at least 90 days). No need to repeat it.

Just logged in, but the first run says “Auth required”

You logged in, but an already-open session hasn't picked up the new auth yet (especially Codex). Open a new thread or restart the editor, then run again.

There's no slash (/) at all in Codex

That's normal. Codex is a tool surface: instead of slash, say “start PurplePrint” and it uses pp_start. Stuck? Diagnose with pp_server_health.

Still stuck? server-health (self-diagnosis: server reachability·source version·auth status + per-tool recovery) and manual (detailed install·auth·troubleshooting) tell you the recovery path directly.