# gptscript を Homebrew, Nix, scoop, winget でインストール

gptscript のインストール経路、実行ファイル、メタデータ、AI エージェント向けセキュリティノートを確認します。

## インストール

```sh
sudo av install brew:gptscript
```

追加のインストールコマンド:

### macOS

- Homebrew (100%):

```sh
brew install gptscript
```

  証拠: local Homebrew formula metadata

### Linux

- Nix (92%):

```sh
nix profile install nixpkgs#gptscript
```

  証拠: nixpkgs package indexes: pkgs/by-name/gp/gptscript/package.nix from https://api.github.com/repos/NixOS/nixpkgs/git/trees/master?recursive=1

### Windows

- Scoop (92%):

```sh
scoop install extras/gptscript
```

  証拠: Scoop official bucket manifest trees: bucket/gptscript.json from https://api.github.com/repos/ScoopInstaller/Extras/git/trees/master?recursive=1

- winget (92%):

```sh
winget install --id gptscript-ai.gptscript -e
```

  証拠: Windows Package Manager source index: gptscript-ai.gptscript from https://cdn.winget.microsoft.com/cache/source.msix

## パッケージ情報

- **パッケージキー:** brew:gptscript
- **パッケージマネージャ:** Homebrew
- **バージョン:** 0.9.9
- **ソース概要:** Develop LLM Apps in Natural Language
- **ホームページ:** <https://docs.gptscript.ai/>
- **リポジトリ:** <https://github.com/gptscript-ai/gptscript>
- **最終更新:** 2026-07-29T16:05:05+02:00
- **生成日時:** 2026-08-03T19:37:03+00:00

## 実行可能ファイル

- gptscript (エイリアス)

## インストール挙動

- Bottle: 利用不可

## バージョンと鮮度

- ページ生成日: 2026-08-03
- マネージャ版: 0.9.9
## プロジェクトの歴史と使われ方

GPTScript is an LLM-oriented scripting framework and CLI from the gptscript-ai project. It packages prompts, local commands, OpenAPI-backed tools, SDK libraries, and retrieval workflows into scripts that can be invoked from the command line.

### プロジェクトの歴史

The public repository began with an initial commit on 2024-01-29. The README describes GPTScript as a framework for allowing large language models to operate against local executables, remote APIs, and retrieval-backed systems with a small prompt-oriented script surface.

The documentation site presents GPTScript as both a scripting language and a CLI workflow runner, with examples for local CLI chat, OpenAPI endpoints, local files, and automated workflows.

### 採用の歴史

GPTScript entered package-manager culture quickly because the official installation path includes Homebrew for macOS and Linux and winget for Windows. The Homebrew formula also appears in the supplied package metadata, alongside Nix, Scoop, and winget identifiers.

The project became part of the early wave of LLM command-line automation tools: useful to developers who wanted prompt-authored workflows without writing a full agent application.

### 使われ方

Typical use is `gptscript` followed by a local script or a GitHub-hosted script. Official examples include running a chat assistant from `github.com/gptscript-ai/llm-basics-demo` and composing tools that call CLIs, web APIs, or files.

The tool needs model credentials for most real workflows; the existing curation records GPTScript's config JSON as both the config and credential storage location.

### パッケージ好きにとっての重要性

For package maintainers, GPTScript is a representative example of post-2023 AI tooling entering ordinary CLI distribution channels. Its packaging spans Homebrew, Nix, Scoop, and winget rather than only an npm, Python, or container ecosystem.

Its command model is also interesting because scripts may reference GitHub-hosted tools directly, making the package itself a launcher and runtime for higher-level LLM workflows.

### タイムライン

- 2024-01-29: Public repository history begins with the initial commit.
- 2024: README copyright names Acorn Labs, Inc. and describes the framework for LLM-operated tools.
- 2026-05: Repository history includes work moving parts of the implementation toward a nanobot repository.

### Related projects

- gptscript-ai/llm-basics-demo is used in the official getting-started command as a runnable demo script.
- OpenAPI-backed services, local CLIs, and retrieval tools are first-class integration targets in the official examples.

### ソース

- <https://docs.gptscript.ai/>
- <https://github.com/gptscript-ai/gptscript>


## セキュリティノート

narrow executable package without higher-risk signals.

- **Geiger リスク:** グリーン / 低
- narrow executable package without higher-risk signals


## Configuration and credential file locations

These source-backed paths show where this package keeps local settings or durable credentials. Automic Vault can use them as review targets for secret scanning, migration, and command approval.


## Configuration files

- Linux: $XDG_CONFIG_HOME/gptscript/config.json
- macOS: ~/Library/Application Support/gptscript/config.json
- Windows: %APPDATA%\Local\gptscript\config.json

## Credential files

- Linux: $XDG_CONFIG_HOME/gptscript/config.json
- macOS: ~/Library/Application Support/gptscript/config.json
- Windows: %APPDATA%\Local\gptscript\config.json
## 他のパッケージマネージャ記録

- Nix - gptscript: normalized package name match | nixpkgs package indexes: pkgs/by-name/gp/gptscript/package.nix from https://api.github.com/repos/NixOS/nixpkgs/git/trees/master?recursive=1
- Scoop - extras/gptscript: normalized package name match | Scoop official bucket manifest trees: bucket/gptscript.json from https://api.github.com/repos/ScoopInstaller/Extras/git/trees/master?recursive=1
- winget - gptscript-ai.gptscript: normalized package name match | Windows Package Manager source index: gptscript-ai.gptscript from https://cdn.winget.microsoft.com/cache/source.msix


## Combined YAML source

View the package source record on GitHub. [combined/gptscript.yml](https://github.com/mxcl/pkgdb/blob/main/combined/gptscript.yml)


## ソース

- pkg.so package database
- Geiger risk classifier
- curated configuration and credential file locations
- curated package history
- pkgdb category and tag curation
- external package-manager database matches
- cross-ecosystem install command graph