# sergent.build

AI proposes. Your code decides.

Propose, Rehearse, Commit.


## User behavior layer

**People set the direction.**

This is the interface where people work: writing, designing, organizing, or
solving a problem. The application records their actions and choices as
structured data.

The user stays in charge of what they want to achieve.

## Agentic layer

**A copilot makes proposals.**

The agent observes the user's actions, choices and recent context, then proposes actions
that support the user's direction and goals. It turns an understanding of the task into
typed proposals over the operations the application exposes.

The model proposes; deterministic software decides which changes can take effect.

## Algorithmic layer

**Deterministic software does the work.**

This layer uses classic software design patterns and engineering practices.
It can be an existing codebase, with a selective Programming Interface that
exposes deterministic operations to the agentic layer above.

The runtime validates the proposed work, rehearses changes in an isolated
copy, then commits when all the required checks pass.

## Three layers. One application.

**Direction, proposals, and controlled effects.**

The three planes assemble into the Sergent architecture. Each layer has a distinct
responsibility, and together they form the Sergent design pattern.

For more information, start with the [Sergent Specification.](https://github.com/sergentbuild/sergent-specification)

If you would like to see the Sergent architecture and design pattern in action, explore the libraries and examples below.

## Find your starting point

The specification, libraries, and examples live on GitHub. Choose your programming language,
then use the library's documentation and its companion examples to get started.

| Language | Library | Examples |
| --- | --- | --- |
| Python | [sergent-py](https://github.com/sergentbuild/sergent-py) | [sergent-py-examples](https://github.com/sergentbuild/sergent-py-examples) |
| Go | [sergent-go](https://github.com/sergentbuild/sergent-go) | [sergent-go-examples](https://github.com/sergentbuild/sergent-go-examples) |
| Rust | [sergent-rs](https://github.com/sergentbuild/sergent-rs) | [sergent-rs-examples](https://github.com/sergentbuild/sergent-rs-examples) |
| TypeScript | [sergent-ts](https://github.com/sergentbuild/sergent-ts) | [sergent-ts-examples](https://github.com/sergentbuild/sergent-ts-examples) |

## Case highlights

### Turn-Based Game

![Orange, blue, and slate chess pieces on a game board.](https://sergent.build/images/cases/turn-based-game.png)

Human vs. AI or AI vs. AI for both fun and AI intelligence evaluation.

[gomoku](https://github.com/sergentbuild/sergent-rs-examples/tree/main/gomoku)

### Creative Content Production

![Two hands collaborate to build a miniature town from orange, blue, and slate blocks.](https://sergent.build/images/cases/creative-content-production.png)

A human drives content creation with a team of AI assistants filling in the gaps.

[ghoul](https://github.com/sergentbuild/sergent-rs-examples/tree/main/ghoul),
[hitchcock](https://github.com/sergentbuild/sergent-py-examples/tree/main/hitchcock)

### Deep Research

![A geometric figure rests its chin on one hand in thought.](https://sergent.build/images/cases/deep-research.png)

Break a complex subject down and then iterate on all the elements to create a full answer.

[deep think](https://github.com/sergentbuild/sergent-py-examples/tree/main/deep_think),
[graph builder](https://github.com/sergentbuild/sergent-py-examples/tree/main/graph_builder)

### Secure Coding Agent

![A laptop displays a code notebook and a plotted graph beside books and a mug.](https://sergent.build/images/cases/secure-coding-agent.png)

AI writes code in a secure sandbox with a built-in adversarial review process.

[coder](https://github.com/sergentbuild/sergent-py-examples/tree/main/coder)

### System Programming

![A desktop computer displays a Linux terminal beside a keyboard, disks, books, and a mug.](https://sergent.build/images/cases/system-programming.png)

Add agentic decision-making to system programs where reliability and portability matter.

[cog](https://github.com/sergentbuild/sergent-rs-examples/tree/main/cog)

### In-Browser LLM

![A browser window displays a page with orange, blue, and slate content blocks.](https://sergent.build/images/cases/in-browser.png)

Run an agentic application in the browser without any external model providers.

[grammary-x](https://github.com/sergentbuild/sergent-ts-examples/tree/main/grammary-x)

Coding agents can read [this page as plain text](https://sergent.build/llms.txt), then follow the
same repository links. Installation and API details belong to each library's
documentation.