# humanbound を Homebrew でインストール
humanbound 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:humanbound
```
追加のインストールコマンド:
### macOS
- Homebrew (100%):
```sh
brew install humanbound
```
証拠: local Homebrew formula metadata
## パッケージ情報
- **パッケージキー:** brew:humanbound
- **パッケージマネージャ:** Homebrew
- **バージョン:** 2.8.0
- **ソース概要:** Adversarial security testing engine, SDK, and CLI for AI agents
- **ホームページ:**
- **最終更新:** 2026-07-31T00:17:01Z
- **生成日時:** 2026-08-03T19:37:03+00:00
## 実行可能ファイル
- hb (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: 2.8.0
## プロジェクトの歴史と使われ方
Humanbound is an open-source adversarial testing engine, Python SDK, and hb command-line interface for AI agents. It drives multi-turn attacks against live agent endpoints, evaluates failures, and can turn findings into runtime guardrail rules.
### プロジェクトの歴史
The official Humanbound repository was created in February 2026 and is maintained by the verified Humanbound GitHub organization. The project combines an Apache-licensed local engine with an optional hosted platform, allowing initial testing without a login while preserving a path to managed posture tracking.
The official release notes describe a consolidation around the humanbound PyPI package: the earlier humanbound-cli package was yanked, and the CLI and public SDK now ship together. The same repository also integrates with the sibling humanbound-firewall package for a test-to-guardrail workflow.
### 採用の歴史
Humanbound is a young project, so its reliable adoption history is limited. Official materials show distribution through PyPI, Homebrew in the supplied package facts, and an official GitHub Action for security gates in continuous integration.
### 使われ方
Users describe an agent endpoint in a JSON configuration file and optionally define permitted and restricted behavior in a scope file. The hb test command runs attacks locally or through the platform; hb posture, hb logs, and hb report inspect the results.
Local attack generation and judging can use OpenAI, Anthropic, Gemini, Azure OpenAI, Grok, or Ollama. Provider settings resolve from CLI flags, environment variables, and then ~/.humanbound/config.yaml; platform login credentials are stored under ~/.humanbound/.
### パッケージ好きにとっての重要性
Humanbound is notable as an AI-security tool that packages an engine, SDK, and CLI under one Python distribution while keeping the firewall as a composable sibling package. Its optional extras let package users choose the provider engine, firewall runtime, or both.
### タイムライン
- 2026: The official humanbound/humanbound repository was created.
- 2026: The project consolidated the CLI and SDK in the humanbound PyPI package and yanked the older humanbound-cli package.
- 2026: The official release feed reached the v2.7 series.
### Related projects
- humanbound-firewall is the sibling Apache-licensed runtime that deploys rules generated from Humanbound findings.
- humanbound/actions is the official GitHub Action for running Humanbound checks in CI.
### ソース
-
-
-
-
-
- source_facts.package-manager
## セキュリティノート
humanbound に一致するローカルシークレット処理マニフェストは見つかりませんでした。将来の対応で安定したパッケージ URL を使えるよう、Nucleus パッケージメタデータはここに公開されています。
## 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
- Unix: ~/.humanbound/config.yaml
## Credential files
- Unix: ~/.humanbound/
## Combined YAML source
View the package source record on GitHub. [combined/humanbound.yml](https://github.com/mxcl/pkgdb/blob/main/combined/humanbound.yml)
## ソース
- pkg.so package database
- curated configuration and credential file locations
- curated package history
- cross-ecosystem install command graph