Skip to content

Commit 728c75b

Browse files
author
Baur
authored
Merge pull request #48 from ZenGuard-AI/nuradil/colab-for-each-detectors
Created colabs for each detectors
2 parents de6e236 + b0ade67 commit 728c75b

File tree

6 files changed

+1253
-0
lines changed

6 files changed

+1253
-0
lines changed

‎docs/colabs/allowed_topics.ipynb‎

Lines changed: 209 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,209 @@
1+
{
2+
"cells": [
3+
{
4+
"cell_type": "markdown",
5+
"metadata": {
6+
"colab_type": "text",
7+
"id": "view-in-github"
8+
},
9+
"source": [
10+
"<a href=\"https://colab.research.google.com/github/ZenGuard-AI/fast-llm-security-guardrails/blob/main/docs/colabs/allowed_topics.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
11+
]
12+
},
13+
{
14+
"cell_type": "markdown",
15+
"metadata": {
16+
"id": "ouO2X6oz9uNR"
17+
},
18+
"source": [
19+
"# ZenGuard AI: Quickstart"
20+
]
21+
},
22+
{
23+
"cell_type": "markdown",
24+
"metadata": {
25+
"id": "OiMYRN6X-VzA"
26+
},
27+
"source": [
28+
"## Intro\n",
29+
"\n",
30+
"The ZenGuard AI provides ultrafast guardrails to protect your GenAI application from:\n",
31+
"\n",
32+
"* Prompt Attacks\n",
33+
"* Veering off pre-defined topics\n",
34+
"* PII, sensitive info, and keywords leakage\n",
35+
"* Toxicity\n",
36+
"* Etc.\n",
37+
"\n",
38+
"Please, also check out our [open-source Python Client](https://github.com/ZenGuard-AI/fast-llm-security-guardrails?tab=readme-ov-file) for more inspiration.\n",
39+
"\n",
40+
"Here is our main website - https://www.zenguard.ai/\n",
41+
"\n",
42+
"More [Docs](https://docs.zenguard.ai/start/intro/)"
43+
]
44+
},
45+
{
46+
"cell_type": "markdown",
47+
"metadata": {
48+
"id": "installation"
49+
},
50+
"source": [
51+
"## Installation\n",
52+
"\n",
53+
"Using pip:"
54+
]
55+
},
56+
{
57+
"cell_type": "code",
58+
"execution_count": null,
59+
"metadata": {
60+
"id": "rt-akQwe7u5j"
61+
},
62+
"outputs": [],
63+
"source": [
64+
"!pip install zenguard -U"
65+
]
66+
},
67+
{
68+
"cell_type": "markdown",
69+
"metadata": {
70+
"id": "prerequisites"
71+
},
72+
"source": [
73+
"## Prerequisites\n",
74+
"\n",
75+
"Configure an API key:\n",
76+
"1. Navigate to the [Settings](https://console.zenguard.ai/settings)\n",
77+
"2. Click on the **+ Create new secret key**.\n",
78+
"3. Name the key **Quickstart Key**.\n",
79+
"4. Click on the **Add** button.\n",
80+
"5. Copy the key value by pressing the copy icon.\n",
81+
"\n",
82+
"Configure Allowed Topics Detection settings:\n",
83+
"1. Navigate to the [Policy](https://console.zenguard.ai/policy)\n",
84+
"2. In API key tabs select **Quickstart Key**.\n",
85+
"3. Scroll down to **Allowed Topics** section\n",
86+
"4. To enable **Allowed Topics Detection** toggle toggler to on state\n",
87+
"5. Enter `Bitcoin` in input and click **Add** button"
88+
]
89+
},
90+
{
91+
"cell_type": "markdown",
92+
"metadata": {
93+
"id": "code-usage"
94+
},
95+
"source": [
96+
"## Code Usage\n",
97+
"\n",
98+
"Instantiate the ZenGuard AI client with the API Key:"
99+
]
100+
},
101+
{
102+
"cell_type": "markdown",
103+
"metadata": {
104+
"id": "api-key"
105+
},
106+
"source": [
107+
"Paste your API key into the env variable **ZEN_API_KEY**:"
108+
]
109+
},
110+
{
111+
"cell_type": "code",
112+
"execution_count": null,
113+
"metadata": {
114+
"id": "set-env"
115+
},
116+
"outputs": [],
117+
"source": [
118+
"%set_env ZEN_API_KEY=YOUR_API_KEY"
119+
]
120+
},
121+
{
122+
"cell_type": "code",
123+
"execution_count": null,
124+
"metadata": {
125+
"id": "zenguard-init"
126+
},
127+
"outputs": [],
128+
"source": [
129+
"from zenguard import Credentials, Detector, ZenGuard, ZenGuardConfig\n",
130+
"import os\n",
131+
"from pprint import pprint\n",
132+
"\n",
133+
"config = ZenGuardConfig(credentials=Credentials(api_key=os.environ.get(\"ZEN_API_KEY\")))\n",
134+
"zenguard = ZenGuard(config=config)\n"
135+
]
136+
},
137+
{
138+
"cell_type": "markdown",
139+
"metadata": {
140+
"id": "detect-injection"
141+
},
142+
"source": [
143+
"## Detect Allowed Topics"
144+
]
145+
},
146+
{
147+
"cell_type": "code",
148+
"execution_count": null,
149+
"metadata": {
150+
"id": "prompt-injection-detection"
151+
},
152+
"outputs": [],
153+
"source": [
154+
"message = \"Tell about Bitcoin\"\n",
155+
"response = zenguard.detect(detectors=[Detector.ALLOWED_TOPICS], prompt=message)\n",
156+
"if response.get(\"is_detected\") is True:\n",
157+
" pprint(\"Allowed Topics detected.\")\n",
158+
"else:\n",
159+
" pprint(\"Allowed Topics not detected\")\n",
160+
"pprint(response)"
161+
]
162+
},
163+
{
164+
"cell_type": "markdown",
165+
"metadata": {},
166+
"source": [
167+
"* `is_detected(boolean)`: Indicates whether a allowed topics was detected in the provided message. In this example, it is False.\n",
168+
"* `score(float: 0.0 - 1.0)`: A score representing the likelihood of the detected allowed topics. In this example, it is 0.0.\n",
169+
"* `sanitized_message(string or null)`: For the allowed topics detector this field is null.\n",
170+
"* `latency(float)`: Server-side latency of the request.\n",
171+
"\n",
172+
"**Error Codes:**\n",
173+
"\n",
174+
"* `401 Unauthorized`: API key is missing or invalid.\n",
175+
"* `400 Bad Request`: The request body is malformed.\n",
176+
"* `500 Internal Server Error`: Internal problem, please escalate to the team."
177+
]
178+
},
179+
{
180+
"cell_type": "markdown",
181+
"metadata": {},
182+
"source": [
183+
"## More examples\n",
184+
"\n",
185+
" * [Detect PII](https://docs.zenguard.ai/detectors/pii/)\n",
186+
" * [Detect Allowed Topics](https://docs.zenguard.ai/detectors/allowed-topics/)\n",
187+
" * [Detect Banned Topics](https://docs.zenguard.ai/detectors/banned-topics/)\n",
188+
" * [Detect Keywords](https://docs.zenguard.ai/detectors/keywords/)\n",
189+
" * [Detect Secrets](https://docs.zenguard.ai/detectors/secrets/)\n",
190+
" * [Detect Toxicity](https://docs.zenguard.ai/detectors/toxicity/)"
191+
]
192+
}
193+
],
194+
"metadata": {
195+
"colab": {
196+
"provenance": [],
197+
"toc_visible": true
198+
},
199+
"kernelspec": {
200+
"display_name": "Python 3",
201+
"name": "python3"
202+
},
203+
"language_info": {
204+
"name": "python"
205+
}
206+
},
207+
"nbformat": 4,
208+
"nbformat_minor": 0
209+
}

0 commit comments

Comments
 (0)