> 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/overview.md).

# Overview

Meet ProtoBot: your AI design assistant inside Protoboard that can search, build, validate, and manage hardware designs through natural language.

ProtoBot is Protoboard's built-in AI assistant. It lives in the left sidebar and gives you conversational access to over 100 tools for searching parts, wiring connections, running validation, organizing the board, and managing your BOM, all through natural language.

Open it by clicking the **animated bot icon** in the leftmost **Activity Bar**. ProtoBot is always scoped to the board you currently have open; switch boards and your chat context switches too.

***

## What ProtoBot Can Do

{% tabs %}
{% tab title="Search & Build" %}

* Search the **ProtoPart library** with semantic search ("find a 3.3V I2C temperature sensor")
* Place, move, duplicate, rename, and remove parts
* Create custom parts, with web search to populate real component specs
* Fetch datasheet information for components
  {% endtab %}

{% tab title="Wire & Allocate" %}

* Create harnesses and link interfaces between parts
* Auto-link interfaces and auto-wire functions for standard protocols (I2C, SPI, UART)
* Allocate and auto-allocate pin resources, per function or for an entire harness
  {% endtab %}

{% tab title="Validate & Fix" %}

* Run full board DRC and explain each error in plain language
* Check compatibility between two parts before you wire them
* Get capacity and utilization status
* Suggest and apply specific fixes
  {% endtab %}

{% tab title="Plan & BOM" %}

* Manage board name, description, notes, requirements, tasks, and tags
* Add, update, and resolve review comments
* View BOM items and cost estimates, refresh supplier pricing, manage supplier records
  {% endtab %}

{% tab title="Layout & Research" %}

* Detect layout issues, auto-arrange parts, auto-route harnesses, find empty space
* Search the web to verify component specifications
* Search the Protoboard documentation
  {% endtab %}
  {% endtabs %}

Every response includes **expandable tool-call cards** showing exactly what ProtoBot did on your board, and a collapsible **extended thinking** view of its reasoning.

***

## Choosing a Model

ProtoBot runs on OpenAI models. Pick one from the **model selector** at the bottom of the chat input:

| Model       | Best For                                                          |
| ----------- | ----------------------------------------------------------------- |
| **gpt-5.4** | The default. Fast and capable; handles most design tasks well     |
| **gpt-5.5** | The most capable option for the hardest multi-step problems (Pro) |

More capable models cost more energy per interaction. The selector shows each model's context window and relative cost; see [Usage, Energy, and Privacy](/protobot/usage-and-energy.md).

### Reasoning effort

Next to the model selector is a **reasoning-effort** control with four settings: **Off**, **Low**, **Medium**, and **High**. It sets how much the model thinks before it acts. **Medium** is the default. Raise it for tricky, multi-step problems (DRC diagnosis, cross-domain compatibility); lower it to save energy on routine work. The control reads **Off** and is disabled for models that do not support reasoning. Your choice is remembered per account.

***

## Where ProtoBot Excels

* **Finding compatible parts:** semantic library search from a plain-language description
* **Wiring:** auto-linking interfaces and allocating resources for standard protocols
* **DRC:** running checks, explaining errors, and suggesting concrete fixes
* **Bulk work:** multi-step workflows like "set up a motor chain end to end"
* **Research:** verifying specs against the web before creating custom parts

## Where Human Review Matters

ProtoBot is a powerful assistant, not a replacement for engineering judgment. Apply your own review for:

| Trust Level                     | When to Apply                            | Examples                                                               |
| ------------------------------- | ---------------------------------------- | ---------------------------------------------------------------------- |
| **Trust and verify**            | Standard tasks, well-known components    | Auto-linking I2C, placing common parts, running DRC                    |
| **Trust but double-check**      | Multi-step workflows, cost-related tasks | BOM estimates, auto-arranging layout                                   |
| **Review carefully**            | Edge cases and unusual configurations    | Custom part creation, cross-domain (thermal + electrical) interactions |
| **Always verify independently** | Safety-critical, procurement, compliance | Final BOM ordering, regulated designs, timing-critical signals         |

ProtoBot may miss datasheet edge cases (timing constraints, pull-up requirements, errata), and it reasons about each engineering domain individually rather than modeling cross-domain effects. Cross-reference datasheets for critical connections, and verify pricing and lead times with your supplier before ordering.

{% hint style="info" %}
When in doubt, ask ProtoBot to **explain its reasoning**, then check the tool-call cards to see exactly what changed.
{% endhint %}

## What ProtoBot Cannot Do

* **Physical simulation:** no SPICE, thermal, or mechanical simulation
* **PCB layout:** Protoboard operates at the system level, not the trace level
* **Manufacturing files:** no Gerber or pick-and-place output
* **Live pricing:** BOM prices come from cached supplier data, not live quotes
* **Account management:** ProtoBot cannot change your plan, billing, or settings

***

## Related pages

* [**Using ProtoBot**](/protobot/how-to-use.md): getting started, common workflows, and prompt patterns
* [**Usage, Energy, and Privacy**](/protobot/usage-and-energy.md): how energy works and where your data goes
* [**Agent Bridge (MCP)**](/agent-bridge-mcp/overview.md): connect your own external agent instead


---

# 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/overview.md?ask=<question>&goal=<endgoal>
```

`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.