Your Own AI Build

An offline-first AI agent for your terminal. It runs as a full-screen TUI that understands your codebase, edits files, executes shell commands, and manages long-running tasks - interactively, headlessly for scripting/CI, or embedded in editors and apps via the Agent Client Protocol (ACP).

Your projects, your models, your machine - nothing leaves unless you say so.

• Local-first by default. Out of the box it talks to a local OpenAI-compatible inference server (the Your Own AI desktop app, llama.cpp, Ollama, or any self-hosted endpoint). No account, no sign-in, no cloud round trip.
• No phone-home, verifiably. The cloud upload, telemetry, and auto-update code paths from the upstream project have been removed from the source, not just switched off. Read the diff.
• Bring any model. Local servers, or any OpenAI- or Anthropic-compatible API, configured per model in config.toml. Coding is the first surface, not the ceiling.
• Your data stays put. Config, sessions, and skills live in ~/.your-own-ai-build/. Scanning other tools' config trees (Claude, Cursor, Codex) is off by default and opt-in per surface via the [compat] config section.

Quickstart

Build from source, then point the agent at your local inference server.

Requirements:

• Rust - the toolchain is pinned by rust-toolchain.toml; rustup installs it automatically on first build.
• protoc - proto codegen uses protoc from PATH (or $PROTOC). Alternatively install DotSlash so the hermetic bin/protoc wrapper can download and run it.
• macOS and Linux are supported build hosts; Windows builds are best-effort.

cargo build -p xai-grok-pager-bin --release   # binary: target/release/your-own-ai-build

Configure your model in ~/.your-own-ai-build/config.toml:

[model.my-local-model]
model = "<model id your server reports>"   # see GET /v1/models on your server
base_url = "http://localhost:11435/v1"     # Your Own AI desktop app default
api_key = "local"
api_backend = "chat_completions"
context_window = 32768

[models]
default = "my-local-model"

Models need a context window of at least 16k tokens to work well; 32k or more is recommended.

Then run your-own-ai-build in a project directory. No sign-in step: with a local or custom endpoint there is nothing to authenticate against unless your provider requires an API key.

For editor or app integration, your-own-ai-build agent stdio speaks ACP over stdio.

Working with the Your Own AI desktop app

The Your Own AI desktop app serves an OpenAI-compatible endpoint on localhost:11435 for both its offline and online models. The agent's model list auto-populates from that server, and each of your AIs appears as a selectable model (use the <name>:agent variant for tool-calling work).

Documentation

The user guide ships with the pager crate: crates/codegen/xai-grok-pager/docs/user-guide/ - getting started, keyboard shortcuts, slash commands, configuration, custom models, theming, MCP servers, skills, plugins, hooks, headless mode, sandboxing, and more.

Repository layout

Path	Contents
crates/codegen/xai-grok-pager-bin	Composition-root package; builds the your-own-ai-build binary
crates/codegen/xai-grok-pager	The TUI: scrollback, prompt, modals, rendering
crates/codegen/xai-grok-shell	Agent runtime + leader/stdio/headless entry points
crates/codegen/xai-grok-tools	Tool implementations (terminal, file edit, search, ...)
crates/codegen/xai-grok-workspace	Host filesystem, VCS, execution, checkpoints
crates/codegen/...	The rest of the CLI crate closure (config, MCP, markdown, sandbox, ...)
crates/common/, crates/build/, prod/mc/	Small shared leaf crates pulled in by the closure
third_party/	Vendored upstream source (Mermaid diagram stack)

Internal crate names keep the upstream xai-grok-* prefix so that periodic merges from upstream stay tractable.

Important

The root Cargo.toml (workspace members, dependency versions, lints, profiles) is generated upstream - treat it as read-only. Prefer editing per-crate Cargo.toml files.

Development

cargo check -p <crate>        # always target specific crates; full-workspace builds are slow
cargo test -p xai-grok-config # per-crate tests
cargo clippy -p <crate>       # lint config: clippy.toml at the repo root
cargo fmt --all               # rustfmt.toml at the repo root

License and attribution

First-party code in this repository is licensed under the Apache License, Version 2.0 - see LICENSE.

Your Own AI Build is a fork of Grok Build by xAI, used under the Apache License 2.0. This fork's principal modifications: rebranding, local-first default endpoints, removal of cloud upload/telemetry/auto-update code, and opt-in (rather than automatic) discovery of other tools' configuration. This repository periodically merges upstream's published code.

Third-party and vendored code remains under its original licenses. See:

• THIRD-PARTY-NOTICES - crates.io / git dependencies, bundled UI themes, and in-tree source ports (including openai/codex and sst/opencode tool implementations)
• crates/codegen/xai-grok-tools/THIRD_PARTY_NOTICES.md - crate-local notice for the codex and opencode ports (license texts + Apache section 4(b) change notice)
• third_party/NOTICE - vendored Mermaid-stack index