> For the complete documentation index, see [llms.txt](https://docs.protoboard.xyz/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.protoboard.xyz/protobot/usage-and-energy.md). # Usage, Energy, and Privacy How ProtoBot energy (Wh) works, tips for getting more out of it, and what happens to your data when you use AI features. ## Energy (Wh) ProtoBot usage is measured in **Energy**, using **Watt-hours (Wh)** as a themed credit unit. Wh is a product credit unit within Protoboard; it does not represent literal electrical power. Every plan includes a daily energy allowance that **refreshes every 24 hours** from your first message. The **battery icon** in the ProtoBot panel shows your current balance. When energy runs out, ProtoBot pauses until your allowance refreshes; the rest of Protoboard (canvas, parts library, DRC, BOM, sharing) keeps working. {% hint style="info" %} Your current allowance and remaining balance are always visible in the app. Allowance amounts are not listed here because they change as the product evolves. {% endhint %} ### What consumes energy Three things drive cost: 1. **Model choice.** More capable models cost more per interaction: **gpt-5.4** (the default) is the efficient option, and **gpt-5.5** costs more. The model selector shows relative pricing. 2. **Reasoning effort.** The reasoning-effort control next to the model selector sets how much the model thinks before acting. Higher effort handles hard problems better but uses more energy; Medium is the default, and Low or Off keep routine work cheap. 3. **Task size.** A short question ("What's on my board?") costs little; a multi-step workflow ("find a regulator, place it, wire it, run DRC") chains many tool calls and costs more. ### Getting more out of your energy * **Stay on the default model** for routine tasks; switch up only for genuinely complex workflows, and switch back after. * **Combine related actions** into one message: "Create a harness, auto-link interfaces, and run DRC" costs less than three separate exchanges. * **Be specific.** One well-constrained prompt replaces several vague back-and-forths. See [prompt patterns](/protobot/how-to-use.md#prompt-patterns). * **Use auto-link and auto-allocate** instead of wiring connections one at a time. * **Start a fresh session when switching topics.** Long sessions accumulate context that must be processed with every message. *** ## Data and Privacy Using ProtoBot means your design data flows through an AI provider. Here is exactly what that involves. ### What gets sent When you message ProtoBot, the request to the model includes your **board data** (parts, interfaces, harnesses, allocations, BOM, name, notes, requirements, tasks) and your **conversation** (messages, responses, tool inputs and outputs). There is no way to use ProtoBot without sending board data to the AI provider. ### Where it goes ProtoBot runs on **OpenAI** models, so requests are processed by OpenAI's API servers. Sessions and messages are also stored by Protoboard so your history persists. For provider-side handling, see [OpenAI's API data usage policies](https://openai.com/policies/api-data-usage-policies). ### What not to include {% hint style="danger" %} Never paste credentials, API keys, or passwords into the chat. They would be sent to the AI provider and stored in your session history. {% endhint %} Also avoid: personally identifiable information about end users, trade secrets, regulated data (HIPAA, ITAR, EAR), and financial details. Technical content (constraints, part numbers, protocols, pin mappings, BOM data) is what ProtoBot is designed for. ### Sensitive projects * **Typical prototypes:** component specs and connection topology are not inherently sensitive; focus your review on correctness rather than exposure. * **Proprietary designs:** use ProtoBot for the standard portions (power supplies, common sensor interfaces) and handle trade-secret subsystems manually. Check your organization's policy on third-party AI tools. * **Regulated industries:** designs under HIPAA, ITAR, ISO 13485, DO-178C, or similar frameworks may restrict third-party AI processing. Consult your compliance team first. The [Privacy Policy](/legal/privacy-policy.md) and [Terms of Service](/legal/terms.md) are the definitive documents governing data use. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.protoboard.xyz/protobot/usage-and-energy.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.