Описание

Рамки управления безопасностью для агентного ИИ. Контролируйте, аудируйте и обеспечивайте выполнение политик безопасности в операциях агентов ИИ. Предотвращайте внедрение вредоносных команд, утечку данных и несанкционированное использование инструментов.

Установка

go install github.com/cisco-ai-defense/defenseclaw@latest

README

     ____         ____                       ____  _
    / __ \  ___  / __/___   ___   ___  ___  / ___|| | __ _ __      __
   / / / / / _ \/ /_// _ \ / _ \ / __|/ _ \| |    | |/ _` |\ \ /\ / /
  / /_/ / /  __/ __//  __/| | | |\__ \  __/| |___ | | (_| | \ V  V /
 /_____/  \___/_/   \___/ |_| |_||___/\___| \____||_|\__,_|  \_/\_/

DefenseClaw

Security governance for OpenClaw and agentic AI runtimes.
Scan capabilities before use, inspect runtime traffic, and export durable audit evidence.

DefenseClaw combines a Python operator CLI, a Go gateway, connector hooks, policy, scanners, and observability exporters. It is an enforcement and evidence layer; it does not prove that an agent, model interaction, or third-party capability is risk-free.

Documentation

The DefenseClaw documentation website is the source of truth for installation, setup, configuration, commands, and operator workflows:

Topic	Canonical guide
Install	Install DefenseClaw
First run	Quickstart
Upgrade	Upgrade
Windows	Native Windows
Connectors	Connector compatibility
Guardrails	Guardrail setup
Configuration	Configuration reference
CLI	CLI reference
Observability	Observability

Repository Markdown is limited to contributor guidance, implementation contracts, package-local notes, generated schema references, test fixtures, and historical design records. Start at docs/README.md.

Source development

Source targets are contributor tooling. They are not an installation or upgrade path for a release-managed host; use the website guides above for those tasks. The checked-in toolchain contracts are Python >=3.10,<3.14 (pyproject.toml) and Go 1.26.4 (go.mod). CI exercises the TypeScript components with Node.js 24.

git clone https://github.com/cisco-ai-defense/defenseclaw.git
cd defenseclaw
make build
make test
make check
make lint

make build produces checkout artifacts without publishing them or changing managed installation state; its pycli step may create or update the shared repository-local .venv. Run make help before using state-changing source targets. All local source targets share that one locked, test-ready .venv; there is no separate production-versus-development Python environment to select. make test, make check, and make py-lint bootstrap that environment when needed. make all intentionally rebuilds and activates the current checkout for local development; the lower-level make install, make dev-install, and scripts/install-dev.sh targets enforce source-ownership rules and are not an upgrade path. Release-managed installations use the release-owned scripts/upgrade.sh or scripts/upgrade.ps1 resolver described on the upgrade page.

The principal source areas are:

Area	Path
Python CLI and TUI	`cli/defenseclaw/`
Go gateway commands	`cmd/`
Go implementation packages	`internal/`
OpenClaw extension	`extensions/defenseclaw/`
Policy bundles	`policies/`
Versioned schemas	`schemas/`
Release and developer automation	`scripts/`
Tests and fixtures	`test/` and component-local `_test.` files

See CONTRIBUTING.md for the pull-request workflow and SECURITY.md for private vulnerability reporting.

License

Apache-2.0. See LICENSE and NOTICE.

Теги

securitygovernanceauditprompt-injection

DefenseClaw — AI Security Governance