See what your AI product costs per customer, per task, and per successful outcome, in one dashboard built from your LLM usage logs. It also shows model mix, cache hit rate, daily cost, and gross margin when you add revenue by customer. One offline HTML file; prices come from a table you edit, and it needs no provider API keys.
Stop runaway AI spend before it eats your margin, with per-customer spend caps, spike alerts, and limits on runaway agent runs. It sets the limits from your own usage, checks usage against them with month-end forecasts and failure rate alerts, and writes a runbook for each alert. Runs on usage files and exits with code 2 on critical alerts, for cron or CI.
Check whether you leaked API keys or secrets before you push, open-source, or launch your app. Finds .env files git tracks or kept in history, server keys exposed to the browser through NEXT_PUBLIC_ or VITE_, Stripe, AWS, OpenAI, Anthropic, and Supabase key shapes, and hardcoded passwords, then walks you through fixing and rotating them. Reports names and redacted hints, never the values.
See where your money goes by sorting a bank or credit card CSV export into spending categories. Uses a plain rules file you own and edit, so the same export always gives the same answer, and writes a categorized CSV, spend by category and month, and unknown merchants with suggested rules. Runs offline; good for budgets and month-end close.
Find out which customers your AI product loses money on, with gross margin per customer and month. It puts revenue next to LLM cost and allocated hosting and other cost of revenue, flags negative and below-target customers, names the task driving their cost, and gives the price each would need to hit your target margin.
Turn Anthropic and OpenAI usage exports, or your app's token logs, into one clean file of token usage and cost by customer and task. It maps API keys and projects to customers, prices tokens from a table you edit, avoids counting cached tokens twice, and lists anything it could not map or price. Works offline and never writes API key values to any file.
Try per seat, per task, credits, or hybrid pricing for your AI product on last month's real usage before you change a price. See revenue and gross margin after LLM cost under each model, and which customers' bills would jump or drop. A planning tool: it never changes prices or contacts customers.
