Superwall ai

content/docs/agents/automations.mdx

@@ -0,0 +1,59 @@
+---
+title: "Automation"
+description: "Schedule recurring prompts for experiment analysis, monitoring, reporting, and follow-up work."
+---
+
+Automations send a saved prompt to a chat on a schedule. Use them for recurring analysis and operational checks that should happen without someone remembering to ask. **Click** on **+ Create Automation** to get started:
+
+<Frame>
+  <img
+    src="/docs/images/agents_new_automation.jpg"
+    alt="The Superwall Agents new automation dialog for scheduling a recurring prompt"
+  />
+</Frame>
+
+Set the prompt and schedule, and the agent will run the prompt at that time, and you can visit the results once they do in your chat history. Automations themselves are attached to chats. The scheduled prompt is sent back into that chat, so the result stays with the surrounding context. Some common automation requests might be:
+
+- Weekly experiment readout.
+- Daily campaign anomaly check.
+- Monday paywall opportunity report.
+- Trial conversion and billing issue review.
+- Weekly SDK integration health check.
+- Webhook delivery summary.
+- Churn or cancellation timing report.
+
+### Create an automation from chat
+
+You can also ask the agent to schedule a recurring prompt:
+
+```text
+Every Monday at 9 AM America/Chicago, analyze the previous week's experiment results and suggest the next three paywall tests.
+```
+
+The agent can configure the chat automation for you when the schedule is clear.
+
+### Write automation prompts carefully
+
+Automated prompts should include enough context to run unattended:
+
+```text
+Analyze the previous 7 days of onboarding campaign performance for the selected Superwall organization. Compare paywall conversion, trial conversion, paid conversion, and ARPU to the prior 7 days. Create charts, summarize notable changes, and recommend three experiments for the next week.
+```
+
+Avoid prompts that depend on follow-up questions:
+
+```text
+Analyze the experiment we talked about.
+```
+
+### Timezones
+
+Automations use the timezone saved with the schedule. If the exact business timezone matters, name it in the prompt and verify it in the automation dialog.
+
+### Limitations
+
+Automations run without a person present. During automated turns, the agent cannot use interactive question or form tools. If the agent needs credentials, organization context, files, or a connected integration, configure those before the automation runs.
+
+### Billing
+
+Scheduled chats require an active Superwall Agents subscription when they run through the hosted control plane. If billing is required, Superwall Agents will show a billing message instead of running the scheduled turn.

content/docs/agents/billing-and-usage.mdx

@@ -0,0 +1,52 @@
+---
+title: "Billing and Usage"
+description: "Review Superwall Agents usage, subscription state, provider costs, and billing controls."
+---
+
+Superwall Agents uses model providers, web search, hosted infrastructure, automations, and webhooks. Superwall Agents includes usage and billing pages so you can see activity and manage your subscription.
+
+<Frame>
+  <img
+    src="/docs/images/agents_settings_billings.jpg"
+    alt="The Settings Billing page for managing the Superwall Agents subscription"
+  />
+</Frame>
+
+### Usage
+
+Open **Settings -> Usage** to review usage for the last 30 days. Usage updates periodically and is grouped by series, such as model and web search usage.
+
+Use this page to understand:
+
+- Which capabilities are driving usage.
+- How usage changed over time.
+- Which series contributed most in the last 30 days.
+
+### Billing
+
+Open **Settings -> Billing** to manage your Superwall Agents subscription.
+
+The Billing page shows:
+
+- Current subscription state.
+- Current plan price.
+- Included monthly usage.
+- Overage terms.
+- Stripe billing controls.
+
+Use the billing page as the source of truth for current pricing and terms.
+
+### When billing is required
+
+Billing may be required for hosted provider usage and unattended workflows, including:
+
+- Managed model provider usage.
+- Managed web search usage.
+- Scheduled chats.
+- Incoming webhooks.
+
+If billing is required, the agent will show a billing message instead of continuing the turn.
+
+### Usage by chat
+
+Some usage views are scoped to chats. Use chat-level usage when you want to understand the cost of a specific analysis, automation, or investigation.

content/docs/agents/create-an-agent.mdx

@@ -0,0 +1,85 @@
+---
+title: "Start a Chat"
+description: "Start a Superwall Agents chat, attach files, and understand where chat state lives."
+---
+
+The main unit of work in Superwall Agents is a chat. A chat keeps its own messages, model and reasoning settings, selected Superwall organization, automation, webhook configuration, attachments, generated files, and transcript history.
+
+<Frame>
+  <img
+    src="/docs/images/agents_welcome_page.png"
+    alt="The Superwall Agents welcome page with the composer, organization selector, model selector, reasoning selector, integrations, and suggested prompts"
+  />
+</Frame>
+
+### Chat about focused actions, results, or experiments
+
+It helps to create one chat per coherent workflow, such as:
+
+- One experiment readout.
+- One SDK integration review.
+- One paywall iteration.
+- One weekly KPI report.
+- One webhook or automation investigation.
+
+Try to avoid mixing unrelated investigations in the same chat. However, the agent _can_ use prior chat history as context, so focused threads produce better answers and analysis.
+
+### Charts and other data visualizations
+
+Often, the agent will create charts to demonstrate its findings:
+
+<Frame>
+  <img
+    src="/docs/images/agents_charts.png"
+    alt="A Superwall Agents chat showing generated charts and analysis output"
+  />
+</Frame>
+
+You can download the charts and expand them fullscreen by using the buttons in the top right of each chart or data visualization.
+
+### Attach files
+
+The chat composer supports file attachments. Use attachments when you want the agent to analyze exports, screenshots, CSV files, logs, or design references. To attach them, simply drag and drop them on top of the chat.
+
+Examples:
+
+- Upload an experiment export and ask for a readout.
+- Upload a screenshot of a paywall and ask for copy or layout experiments.
+- Upload logs and ask the agent to find SDK integration issues.
+
+You can attach up to 10 files at a time. Uploaded files are made available to the hosted machine so the agent can use them during the turn.
+
+### Queuing and steering
+
+If you send a message while the agent is already running, the message is queued. Use this when you want to add follow-up context without waiting.
+
+If the running turn is going in the wrong direction, steer it with a new instruction. Steering interrupts the current turn and restarts with your updated direction.
+
+### Generated files
+
+Agents can generate files, show charts, and display artifacts in Superwall Agents. Use this for reports, CSV summaries, markdown writeups, scripts, implementation patches, segment comparisons, time series, experiment variants, revenue breakdowns, and outlier analysis.
+
+Ask for a specific artifact:
+
+```text
+Create a markdown report summarizing the latest experiment, including charts, interpretation, and the next three tests we should run.
+```
+
+Generated files are available from **Files** in the left sidebar, so you can reopen or download them after the chat finishes.
+
+<Frame>
+  <img
+    src="/docs/images/agents_file_download.png"
+    alt="The Superwall Agents Files tab showing a generated file ready to open or download"
+  />
+</Frame>
+
+### Transcripts
+
+The chat keeps transcript history so later turns can use earlier context.
+
+When you need prior decisions or historical context, ask the agent to search its transcripts:
+
+```text
+Search our prior transcripts for decisions about onboarding paywall experiments, then summarize what we already tried and what we should do next.
+```

content/docs/agents/files.mdx

@@ -0,0 +1,32 @@
+---
+title: "Files"
+description: "Browse files available on the active hosted machine and open them for agent work."
+---
+
+The **Files** page shows files available on the active hosted machine. Use **Files** when you want to inspect your agent's hosted environment's files you've created through chats.
+
+<Frame>
+  <img
+    src="/docs/images/agents_file_download.png"
+    alt="The Superwall Agents Files tab showing a generated file ready to open or download"
+  />
+</Frame>
+
+All files created are listed, and can be opened in-place. The navigator includes options to refresh all files, and in some cases — files will have their own folder and directory. You can navigate through each of them here.
+
+### Previews
+
+File previews use the same attachment preview system as chat. Images, videos, text, markdown, PDFs, JSON, code, and common tabular files can be inspected from the UI when the hosted machine can serve them.
+
+<Frame>
+  <img
+    src="/docs/images/agents_file_preview.jpg"
+    alt="The Superwall Agents file preview view showing a generated file opened from Files"
+  />
+</Frame>
+
+If a file cannot be previewed inline, Superwall Agents opens it in a new tab. In the preview, you can download the file or view the raw source response that generated it.
+
+<Warning>
+Treat Files as a hosted machine file browser. Do not place secrets in files you ask the agent to inspect unless the agent truly needs them. Prefer [Settings -> Environment](/agents/settings#environment) for secret values.
+</Warning>

content/docs/agents/hosted-machines.mdx

@@ -0,0 +1,67 @@
+---
+title: "Hosted Machines"
+description: "Manage hosted machines for Superwall Agents work."
+---
+
+Superwall Agents runs from [superwall.ai](https://superwall.ai) against a hosted machine managed by Superwall. The active machine is where chat tools, Files, Terminal, automations, and webhook-triggered work execute.
+
+<Frame>
+  <img
+    src="/docs/images/agents-machines-placeholder.svg"
+    alt="A Machines settings page showing a hosted machine with status, version, activate, update, rename, and delete actions"
+  />
+</Frame>
+
+Image to add: a screenshot of Settings -> Machine showing the active hosted machine, the Add Machine button, the status indicator, and machine actions.
+
+### When to use hosted machines
+
+Hosted machines are useful for:
+
+- Starting from the browser.
+- Running agent work without relying on your computer.
+- Isolating agent work from your computer.
+- Scheduling automations and webhook-triggered workflows.
+- Keeping Files and Terminal attached to the same remote environment.
+
+Hosted machines have their own filesystem and environment. They cannot see files on your laptop unless those files are uploaded, cloned, or otherwise made available to the machine.
+
+### Machine settings
+
+Open **Settings -> Machine** to:
+
+- Add a hosted machine.
+- Activate a machine.
+- Reconnect to the active machine.
+- Update a hosted machine.
+- Rename a hosted machine.
+- Delete a hosted machine.
+
+The active machine determines where chat tools, Terminal, Files, automations, and webhook-triggered work run.
+
+### Working with code and files
+
+If you want the agent to inspect code, make the repo available to the hosted machine first. Common paths are:
+
+- Connect GitHub.
+- Clone the repository from Terminal.
+- Upload specific files in the chat composer.
+- Use webhooks or integrations to send files into the chat.
+
+Then ask for the specific work:
+
+```text
+Inspect the connected iOS repo and verify our Superwall SDK integration.
+```
+
+For analysis workflows, make the business question explicit:
+
+```text
+Analyze the latest experiment results and generate a weekly report.
+```
+
+### Data and privacy
+
+Hosted machines store their own machine state in the hosted environment, while account state lives in the hosted control plane.
+
+Do not assume a hosted machine has access to a file or secret on your laptop. Upload the file, connect the repository, or add the environment variable in Settings before asking the agent to use it.

content/docs/agents/index.mdx

@@ -0,0 +1,45 @@
+---
+title: "Superwall Agents"
+description: "Use Superwall Agents to analyze experiments, inspect your setup, automate reports, and turn Superwall data into growth work."
+---
+
+Superwall Agents gives you an AI workspace for the growth work that usually takes the most digging: reading experiment results, finding patterns in your data, checking implementation details, and turning what it finds into the next thing to try.
+
+Use Superwall Agents from [superwall.ai](https://superwall.ai). Log in with your existing Superwall account to access all of your organizations and apps.
+
+<Frame>
+  <img
+    src="/docs/images/agents-overview.png"
+    alt="A Superwall Agents workspace showing a chat analyzing experiment results, a chart, and suggested next experiments"
+  />
+</Frame>
+
+### What you can do
+
+- **Analyze experiment results**: Ask what changed, which variant is winning, what segments are driving the result, and what the next test should be when your connected Superwall tooling exposes that data.
+- **Turn analysis into action**: Ask for concrete experiment ideas, paywall changes, campaign adjustments, or implementation checks.
+- **Inspect available code**: Connect GitHub, upload files, or clone a repo onto the hosted machine, then ask the agent to read files, run commands, and help debug SDK integration issues.
+- **Automate recurring work**: Schedule a prompt for a chat, such as a weekly experiment readout or a daily anomaly check.
+- **Connect external systems**: Use webhooks to trigger a chat from another service, and send chat events back to your own tooling.
+- **Generate artifacts**: Create charts, summaries, files, and reports that can be shared with your team.
+
+### Start here
+
+1. **[Workspace tour](/agents/workspace-tour):** Learn what the sidebar, composer, selectors, integrations, suggested prompts, and selected UI states do.
+2. **[Create an agent chat](/agents/create-an-agent):** Start a focused chat, attach context, review generated artifacts, steer running work, and share useful conversations.
+
+### Using agents
+
+3. **[Automation](/agents/automations):** Schedule recurring prompts for reports, monitoring, experiment review, and other work that should happen on a cadence.
+4. **[Webhooks](/agents/webhooks):** Trigger chats from external systems and send chat events back to your own tooling.
+5. **[Sharing Chats](/agents/sharing-chats):** Share a completed conversation with teammates and manage the shared link.
+6. **[Integrations](/agents/integrations):** Connect GitHub, Slack, skills, web search, models, and hosted machine environment variables.
+7. **[Files](/agents/files):** Browse files available on the active hosted machine, open generated reports, and inspect uploaded or cloned context.
+8. **[Terminal](/agents/terminal):** Run commands on the same hosted machine the agent can use.
+9. **[Settings](/agents/settings):** Configure defaults, environment variables, machines, usage, billing, webhooks, keybindings, and account controls.
+
+### Misc
+
+10. **[Billing and usage](/agents/billing-and-usage):** Review usage, understand how work is counted, and manage the subscription attached to your account.
+11. **[Troubleshooting](/agents/troubleshooting):** Resolve common setup, machine, integration, file, webhook, billing, and chat issues.
+12. **[Recipes](/agents/recipes):** Try prompts for experiment analysis, campaign review, implementation checks, recurring reports, and follow-up ideas.