Cost controls & operations PRO
Pro treats your AI budget like production infrastructure: visible, capped, and portable.
Monthly usage limits
Under Settings → Advanced:
| Limit | Behavior |
|---|---|
| Monthly spend limit ($) | Site-wide 30-day cap. When reached, every AI action is blocked with a clear message until the window rolls. 0 disables. |
| Monthly token limit | Same, in tokens. |
| Per-role limits (JSON) | e.g. {"author":{"cost":5,"tokens":100000}} — authors can draft freely but can't outspend their allowance. |
Limits are enforced at the single point every tool passes through — nothing bypasses them, including autopilots and visitor-facing AI.
Pre-run cost estimates
Batch triggers (Bulk Content, Repair All, Automations run-now, autopilot enabling) show a confirmation modal with the estimated cost before anything is spent — based on the item count and your current model's pricing.
Event webhooks
Set a webhook URL (Zapier / Make / n8n compatible) and Auto RepAI POSTs JSON on:
content_generated, form_submitted, automation_item_completed.
Use it to notify Slack, append to a content-calendar sheet, or trigger downstream publishing steps. A separate Slack webhook field posts AI Form submissions directly to a channel.
Settings export/import
Roll your configuration across sites (the agency workflow):
- Settings → Advanced → Export settings (JSON) downloads your configuration — models, defaults, limits, white-label branding, chatbot settings.
- On the next site: Import settings, pick the file, confirm.
Security by design: API keys, Reddit credentials and webhook URLs are never exported and never overwritten by an import — each site keeps its own secrets. Invalid files are rejected with a clear message.