You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/about/overview.md
+5-5Lines changed: 5 additions & 5 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -48,7 +48,7 @@ The NeMo Guardrails library provides multiple approaches to content safety:
48
48
-**LLM self-checking**: Use the LLM itself to check inputs and outputs for harmful content.
49
49
-**NVIDIA safety models**: Integrate with [Llama 3.1 NemoGuard 8B Content Safety](https://build.nvidia.com/nvidia/llama-3_1-nemotron-safety-guard-8b-v3) for robust content moderation.
50
50
-**Community models**: Use [LlamaGuard](../user-guides/community/llama-guard.md), [Fiddler Guardrails](../user-guides/community/fiddler.md), and other community content safety solutions.
51
-
-**Third-party APIs**: Integrate with [ActiveFence](../configure-rails/yaml-schema/guardrails-configuration/built-in-guardrails.md#activefence), [Cisco AI Defense](../user-guides/community/ai-defense.md), and other moderation services.
51
+
-**Third-party APIs**: Integrate with [ActiveFence](../configure-rails/guardrail-catalog.md#activefence), [Cisco AI Defense](../user-guides/community/ai-defense.md), and other moderation services.
52
52
53
53
For practical examples, try the following tutorials:
54
54
@@ -98,7 +98,7 @@ The NeMo Guardrails library supports PII detection through multiple integrations
98
98
99
99
PII detection can be configured to either detect and block content containing PII or to mask PII entities before processing.
100
100
101
-
For more information, refer to the [Presidio Integration](../user-guides/community/presidio.md) and [Sensitive Data Detection section](../configure-rails/yaml-schema/guardrails-configuration/built-in-guardrails.md#presidio-based-sensitive-data-detection) in the built-in Guardrails library.
101
+
For more information, refer to the [Presidio Integration](../user-guides/community/presidio.md) and [Sensitive Data Detection section](../configure-rails/guardrail-catalog.md#presidio-based-sensitive-data-detection) in the Guardrail Catalog.
102
102
:::
103
103
104
104
:::{dropdown} 🤖 Add Agentic Security
@@ -108,7 +108,7 @@ This includes:
108
108
109
109
-**Tool call validation**: Execute rails that validate tool inputs and outputs before and after invocation.
110
110
-**Agent workflow protection**: Integrate with [LangGraph](../integration/langchain/langgraph-integration.md) for multi-agent safety.
111
-
-**Secure tool integration**: Review guidelines for safely connecting LLMs to external resources (refer to [Security Guidelines](../security/guidelines.md)).
111
+
-**Secure tool integration**: Review guidelines for safely connecting LLMs to external resources (refer to [Security Guidelines](../resources/security/guidelines.md)).
112
112
-**Action monitoring**: Monitor detailed logging and tracing of agent actions.
113
113
114
114
Key security considerations for agent systems:
@@ -118,7 +118,7 @@ Key security considerations for agent systems:
118
118
3. Apply execution rails to tool calls.
119
119
4. Monitor agent behavior for unexpected actions.
120
120
121
-
For more information, refer to the [Tools Integration Guide](../integration/tools-integration.md), [Security Guidelines](../security/guidelines.md), and [LangGraph Integration](../integration/langchain/langgraph-integration.md).
121
+
For more information, refer to the [Tools Integration Guide](../integration/tools-integration.md), [Security Guidelines](../resources/security/guidelines.md), and [LangGraph Integration](../integration/langchain/langgraph-integration.md).
122
122
:::
123
123
124
124
:::{dropdown} 🔧 Build Your Own or Use Third-party Guardrail Solutions
@@ -130,7 +130,7 @@ If you have a script or tool that runs a custom guardrail, you can use it in NeM
130
130
131
131
2.**LangChain tool integration**: Register LangChain tools as custom actions. For more information, refer to the [](../integration/tools-integration.md).
132
132
133
-
3.**Third-party API integration**: Integrate external moderation and validation services. For a complete list of supported third-party guardrail services, refer to the [](../configure-rails/yaml-schema/guardrails-configuration/built-in-guardrails.md#third-party-apis) section in the built-in Guardrails library.
133
+
3.**Third-party API integration**: Integrate external moderation and validation services. For a complete list of supported third-party guardrail services, refer to the [Third-Party APIs](../configure-rails/guardrail-catalog.md#third-party-apis) section in the Guardrail Catalog.
134
134
:::
135
135
136
136
:::{dropdown} 🔌 Integrate NeMo Guardrails Library into Your Application
Follow these steps to install the NeMo Guardrails library.
17
17
18
-
## Requirements
18
+
## Prerequisites
19
19
20
20
Verify your system meets the following requirements before installation.
21
21
@@ -46,7 +46,7 @@ Use the following steps to install the NeMo Guardrails library in a virtual envi
46
46
47
47
```bash
48
48
python -m venv .venv
49
-
.venv\Scripts\activate
49
+
source.venv/Scripts/activate
50
50
```
51
51
52
52
:::
@@ -67,25 +67,14 @@ Use the following steps to install the NeMo Guardrails library in a virtual envi
67
67
68
68
This is required to access NVIDIA-hosted models on [build.nvidia.com](https://build.nvidia.com). The tutorials and example configurations ([examples/configs](https://github.com/NVIDIA-NeMo/Guardrails/tree/develop/examples/configs)) in this library include configurations that use NVIDIA-hosted models.
69
69
70
-
## Install from Source
70
+
## Alternative Installation Methods
71
71
72
-
To use the latest development version:
72
+
Install the NeMo Guardrails library from source using pip or Poetry. Choose this method if you want to contribute to the library or use the latest development version.
@@ -124,7 +113,7 @@ You can install the NeMo Guardrails library with optional extra packages to add
124
113
|-------|-------------|
125
114
|`nvidia`| NVIDIA-hosted model integration through [build.nvidia.com](https://build.nvidia.com/)|
126
115
|`openai`| OpenAI-hosted model integration |
127
-
|`sdd`|[Sensitive data detection](../configure-rails/yaml-schema/guardrails-configuration/built-in-guardrails.md#presidio-based-sensitive-data-detection) using Presidio |
116
+
|`sdd`|[Sensitive data detection](../configure-rails/guardrail-catalog.md#presidio-based-sensitive-data-detection) using Presidio |
128
117
|`eval`|[Evaluation tools](../evaluation/index.rst) for testing guardrails |
For more information, refer to [](../run-rails/using-fastapi-server/index.md).
54
59
55
-
4.**Docker deployment**: Deploy guardrails as a containerized service.
56
-
For more information, refer to the [Using Docker Guide](../deployment/using-docker.md).
60
+
- Use the NeMo Guardrails Docker deployment capabilities to deploy guardrails as a containerized service.
61
+
For more information, refer to [](../deployment/using-docker.md).
57
62
58
-
For complete examples and detailed integration patterns, refer to the [examples directory](https://github.com/NVIDIA-NeMo/Guardrails/tree/develop/examples) in the GitHub repository.
63
+
For more examples and detailed integration patterns, refer to the [examples directory](https://github.com/NVIDIA-NeMo/Guardrails/tree/develop/examples) in the NeMo Guardrails GitHub repository.
@@ -57,7 +57,7 @@ You will secure an application LLM and test block prompt injection and jailbreak
57
57
58
58
The [Nemoguard Jailbreak Detect](https://build.nvidia.com/nvidia/nemoguard-jailbreak-detect) model does not use any prompts, so you don't need to create a `prompts.yml` file for this model.
59
59
60
-
For more information about the configuration parameters, refer to the [Configuration Reference](../../configure-rails/yaml-schema/configuration-reference.md).
60
+
For more information about the configuration parameters, refer to the [Configuration Reference](../../configure-rails/configuration-reference.md).
61
61
62
62
## Run the Guardrails chat application
63
63
@@ -165,7 +165,7 @@ This section shows how to run the NVIDIA NemoGuard JailbreakDetect NIM microserv
165
165
166
166
To run the NVIDIA NemoGuard JailbreakDetect NIM in a Docker container, follow these steps:
167
167
168
-
1. Update the `config.yml` file you created earlier to point to a local NIM deployment rather than build.nvidia.com. The following configuration updates the `nim_base_url` to point to `http://localhost:8123`, which tells the NeMo Guardrails toolkit to make requests to the local NIM deployment. The Guardrails configuration must match the NIM Docker container configuration for them to communicate.
168
+
1. Update the `config.yml` file you created earlier to point to a local NIM deployment rather than build.nvidia.com. The following configuration updates the `nim_base_url` to point to `http://localhost:8123`, which tells the NeMo Guardrails library to make requests to the local NIM deployment. The Guardrails configuration must match the NIM Docker container configuration for them to communicate.
169
169
170
170
```yaml
171
171
models:
@@ -249,4 +249,4 @@ To run the NVIDIA NemoGuard JailbreakDetect NIM in a Docker container, follow th
249
249
250
250
- [NVIDIA NemoGuard JailbreakDetect NIM documentation](https://docs.nvidia.com/nim/nemoguard-jailbreakdetect/latest/index.html)
251
251
- [Jailbreak Detection Heuristics](../../user-guides/jailbreak-detection-heuristics/README.md) for detection without a NIM
252
-
- [Configuration Reference](../../configure-rails/yaml-schema/configuration-reference.md) for all configuration options
252
+
- [Configuration Reference](../../configure-rails/configuration-reference.md) for all configuration options
@@ -122,7 +122,7 @@ By following this tutorial, you learn how to configure a set of allowed topics a
122
122
I'd be happy to help you with canceling your subscription. You have a couple of options to do so, and I'll walk you
123
123
through them.
124
124
125
-
[The NeMo Guardrails toolkit responds with instructions and information on subscription cancellations]
125
+
[The NeMo Guardrails library responds with instructions and information on subscription cancellations]
126
126
```
127
127
128
128
## Import the NeMo Guardrails Library in Python
@@ -185,7 +185,7 @@ This section shows how to run the NemoGuard 8B TopicControl model locally while
185
185
186
186
To run the Llama 3.1 NemoGuard 8B TopicControl in a Docker container, follow these steps:
187
187
188
-
1. Update the `config.yml` file you created earlier to point to a local NIM deployment rather than build.nvidia.com. The following configuration adds a `base_url` and `model_name` field under `parameters`, which tells the NeMo Guardrails toolkit to make requests to the `nvidia/llama-3.1-nemoguard-8b-topic-control` model hosted at `http://localhost:8123/v1`. The Guardrails configuration must match the NIM Docker container configuration for them to communicate.
188
+
1. Update the `config.yml` file you created earlier to point to a local NIM deployment rather than build.nvidia.com. The following configuration adds a `base_url` and `model_name` field under `parameters`, which tells the NeMo Guardrails library to make requests to the `nvidia/llama-3.1-nemoguard-8b-topic-control` model hosted at `http://localhost:8123/v1`. The Guardrails configuration must match the NIM Docker container configuration for them to communicate.
189
189
190
190
```yaml
191
191
models:
@@ -270,7 +270,7 @@ To run the Llama 3.1 NemoGuard 8B TopicControl in a Docker container, follow the
- GPU with at least 16GB VRAM (see [Hardware Requirements](https://huggingface.co/nvidia/Nemotron-Content-Safety-Reasoning-4B#hardware-and-software-requirements) on HuggingFace)
47
47
- vLLM installed:
48
48
@@ -218,6 +218,7 @@ prompts:
218
218
```
219
219
220
220
The `reasoning_enabled` variable is automatically passed to prompt templates by the content safety action, based on the `rails.config.content_safety.reasoning.enabled` setting.
Copy file name to clipboardExpand all lines: docs/getting-started/tutorials/nemotron-safety-guard-deployment.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -193,7 +193,7 @@ Follow these steps to prepare the guardrails configuration.
193
193
```console
194
194
> मुझे भारत में तीन आम खाने की चीज़ज़ों के नाम बताओ।
195
195
196
-
[The NeMo Guardrails toolkit responds with information about rice, roti, and dal—common Indian foods]
196
+
[The NeMo Guardrails library responds with information about rice, roti, and dal—common Indian foods]
197
197
```
198
198
199
199
## Import the NeMo Guardrails Library in Python
@@ -259,7 +259,7 @@ This section shows how to run the Nemotron Safety Guard 8B model locally while s
259
259
260
260
To run the Llama 3.1 Nemotron Safety Guard 8B V3 in a Docker container, follow these steps:
261
261
262
-
1. Update the `config.yml` file you created earlier to point to a local NIM deployment rather than build.nvidia.com. The following configuration adds a `base_url` and `model_name` field under `parameters`, which tells the NeMo Guardrails toolkit to make requests to the `nvidia/llama-3.1-nemotron-safety-guard-8b-v3` model hosted at `http://localhost:8123/v1`. The Guardrails configuration must match the NIM Docker container configuration for them to communicate.
262
+
1. Update the `config.yml` file you created earlier to point to a local NIM deployment rather than build.nvidia.com. The following configuration adds a `base_url` and `model_name` field under `parameters`, which tells the NeMo Guardrails library to make requests to the `nvidia/llama-3.1-nemotron-safety-guard-8b-v3` model hosted at `http://localhost:8123/v1`. The Guardrails configuration must match the NIM Docker container configuration for them to communicate.
0 commit comments