Metadata-Version: 2.5
Name: archeguard
Version: 3.1.4
Summary: ArcheGuard Python SDK - AI Security Layer
Project-URL: Homepage, https://archeguard.com
License: Proprietary
Requires-Python: >=3.10
Requires-Dist: httpx>=0.25.0
Requires-Dist: pydantic>=2.0.0
Requires-Dist: pyyaml>=6.0
Provides-Extra: all
Requires-Dist: anthropic>=0.20.0; extra == 'all'
Requires-Dist: openai>=1.0.0; extra == 'all'
Provides-Extra: anthropic
Requires-Dist: anthropic>=0.20.0; extra == 'anthropic'
Provides-Extra: dev
Requires-Dist: numpy>=1.24; extra == 'dev'
Requires-Dist: pytest-asyncio>=0.21; extra == 'dev'
Requires-Dist: pytest>=7.0; extra == 'dev'
Requires-Dist: respx>=0.20; extra == 'dev'
Provides-Extra: openai
Requires-Dist: openai>=1.0.0; extra == 'openai'
Description-Content-Type: text/markdown

# ArcheGuard Python SDK

Client for the ArcheGuard agent, the self-managed AI security sidecar.
The SDK sends prompts, responses and tool calls to an agent you run and
applies its decisions; it does no detection itself. Without a reachable
agent it fails open by default (`fail_open=False` raises
`AgentUnavailableError` instead).

```bash
pip install archeguard
```

```python
from archeguard import guard
from openai import OpenAI

client = guard(OpenAI())  # agent at ARCHEGUARD_URL, default http://localhost:7777
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}],
)
```

Integrations for LangChain, OpenAI Agents, CrewAI, Pydantic AI and
LiveKit live under `archeguard.integrations`. The agent and its licence
come from ArcheGuard: https://archeguard.com
