# Panoptica GenAI Protection SDK
A simple python client SDK for integration with Panoptica GenAI Protection.
___
GenAI Protection is part of Panoptica, a cloud native application protection platform (CNAPP), and provides protection for LLM-backed systems.
Specifically, the GenAI Protection SDK inspects both input and output prompts, flagging those it identifies as likely containing malicious content with a high degree of certainty.
The Python SDK is provided to programmatically integrate your system with our LLM protection software,
enabling you to verify the safety level of processing a user requested prompt before actually processing it.
Following this evaluation, the application can then determine the appropriate subsequent steps based on your policy.
## Installation
```shell
pip install panoptica_genai_protection
```
## Usage Example
Working assumptions:
- You have generated a key-pair for GenAI Protection in the Panoptica settings screen
* The access key is set in the GENAI_PROTECTION_ACCESS_KEY environment variable
* The secret key is set in the GENAI_PROTECTION_SECRET_KEY environment variable
- We denote the call to generating the LLM response as get_llm_response()
GenAIProtectionClient provides the check_llm_prompt method to determine the safety level of a given prompt.
### Sample Snippet
```python
from panoptica_genai_protection.client import GenAIProtectionClient
from panoptica_genai_protection.gen.models import Result as InspectionResult
# ... Other code in your module ...
# initialize the client
genai_protection_client = GenAIProtectionClient()
# Send the prompt for inspection BEFORE sending it to the LLM
inspection_result = genai_protection_client.check_llm_prompt(
chat_request.prompt,
api_name="chat_service", # Name of the service running the LLM
api_endpoint_name="/chat", # Name of the endpoint serving the LLM interaction
sequence_id=chat_id, # UUID of the chat, if you don't have one, provide `None`
actor="John Doe", # Name of the "actor" interacting with the LLM service.
actor_type="user", # Actor type, one of {"user", "ip", "bot"}
)
if inspection_result.result == InspectionResult.safe:
# Prompt is safe, generate an LLM response
llm_response = get_llm_response(
chat_request.prompt
)
# Call GenAI protection on LLM response (completion)
inspection_result = genai_protection_client.check_llm_response(
prompt=chat_request.prompt,
response=llm_response,
api_name="chat_service",
api_endpoint_name="/chat",
actor="John Doe",
actor_type="user",
request_id=inspection_result.reqId,
sequence_id=chat_id,
)
if inspection_result.result != InspectionResult.safe:
# LLM answer is flagged as unsafe, return a predefined error message to the user
answer_response = "Something went wrong."
else:
# Prompt is flagged as unsafe, return a predefined error message to the user
answer_response = "Something went wrong."
```
#### Async use:
You may use the client in async context in two ways:
```python
async def my_async_call_to_gen_ai_protection(prompt: str):
client = GenAIProtectionClient(as_async=True)
return await client.check_llm_prompt_async(
prompt=prompt,
api_name="test",
api_endpoint_name="/test",
actor="John Doe",
actor_type="user"
)
```
or
```python
async def my_other_async_call_to_gen_ai_protection(prompt: str):
async with GenAIProtectionClient() as client:
return await client.check_llm_prompt_async(
prompt=prompt,
api_name="test",
api_endpoint_name="/test",
actor="John Doe",
actor_type="user"
)
```
Raw data
{
"_id": null,
"home_page": null,
"name": "panoptica-genai-protection",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": null,
"author": "marvin-team@cisco.com",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/33/19/f5ac072ab9ec75a64d8421b48f359bf314977b47f040565154165c97b4f2/panoptica_genai_protection-0.1.12.tar.gz",
"platform": null,
"description": "# Panoptica GenAI Protection SDK\nA simple python client SDK for integration with Panoptica GenAI Protection.\n___\nGenAI Protection is part of Panoptica, a cloud native application protection platform (CNAPP), and provides protection for LLM-backed systems.\nSpecifically, the GenAI Protection SDK inspects both input and output prompts, flagging those it identifies as likely containing malicious content with a high degree of certainty. \n\nThe Python SDK is provided to programmatically integrate your system with our LLM protection software,\nenabling you to verify the safety level of processing a user requested prompt before actually processing it.\nFollowing this evaluation, the application can then determine the appropriate subsequent steps based on your policy.\n\n## Installation\n```shell\npip install panoptica_genai_protection\n```\n\n## Usage Example\nWorking assumptions:\n- You have generated a key-pair for GenAI Protection in the Panoptica settings screen\n * The access key is set in the GENAI_PROTECTION_ACCESS_KEY environment variable\n * The secret key is set in the GENAI_PROTECTION_SECRET_KEY environment variable\n- We denote the call to generating the LLM response as get_llm_response()\n\nGenAIProtectionClient provides the check_llm_prompt method to determine the safety level of a given prompt.\n\n\n### Sample Snippet\n\n```python\nfrom panoptica_genai_protection.client import GenAIProtectionClient\nfrom panoptica_genai_protection.gen.models import Result as InspectionResult\n\n# ... Other code in your module ...\n\n# initialize the client\ngenai_protection_client = GenAIProtectionClient()\n\n# Send the prompt for inspection BEFORE sending it to the LLM\ninspection_result = genai_protection_client.check_llm_prompt(\n chat_request.prompt,\n api_name=\"chat_service\", # Name of the service running the LLM\n api_endpoint_name=\"/chat\", # Name of the endpoint serving the LLM interaction\n sequence_id=chat_id, # UUID of the chat, if you don't have one, provide `None`\n actor=\"John Doe\", # Name of the \"actor\" interacting with the LLM service.\n actor_type=\"user\", # Actor type, one of {\"user\", \"ip\", \"bot\"}\n)\n\nif inspection_result.result == InspectionResult.safe:\n # Prompt is safe, generate an LLM response\n llm_response = get_llm_response(\n chat_request.prompt\n )\n\n # Call GenAI protection on LLM response (completion)\n inspection_result = genai_protection_client.check_llm_response(\n prompt=chat_request.prompt,\n response=llm_response,\n api_name=\"chat_service\",\n api_endpoint_name=\"/chat\",\n actor=\"John Doe\",\n actor_type=\"user\",\n request_id=inspection_result.reqId,\n sequence_id=chat_id,\n )\n if inspection_result.result != InspectionResult.safe:\n # LLM answer is flagged as unsafe, return a predefined error message to the user\n answer_response = \"Something went wrong.\"\nelse:\n # Prompt is flagged as unsafe, return a predefined error message to the user\n answer_response = \"Something went wrong.\"\n```\n\n#### Async use:\nYou may use the client in async context in two ways:\n```python\nasync def my_async_call_to_gen_ai_protection(prompt: str):\n client = GenAIProtectionClient(as_async=True)\n return await client.check_llm_prompt_async(\n prompt=prompt,\n api_name=\"test\",\n api_endpoint_name=\"/test\",\n actor=\"John Doe\",\n actor_type=\"user\"\n )\n```\nor\n```python\nasync def my_other_async_call_to_gen_ai_protection(prompt: str):\n async with GenAIProtectionClient() as client:\n return await client.check_llm_prompt_async(\n prompt=prompt,\n api_name=\"test\",\n api_endpoint_name=\"/test\",\n actor=\"John Doe\",\n actor_type=\"user\"\n )\n```\n",
"bugtrack_url": null,
"license": null,
"summary": "Protecting GenAI from Prompt Injection",
"version": "0.1.12",
"project_urls": null,
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "9ddfcd933ad4f787bfe46365d9cf4a5ba9db2225744db685335f06741ee608f3",
"md5": "938f8dbe854d231b9d91cd787fb1b1d4",
"sha256": "c07f17c86c31c8ce38efd2895dd9188465f31f1de5114fd2b45302ad6ffbb487"
},
"downloads": -1,
"filename": "panoptica_genai_protection-0.1.12-py3-none-any.whl",
"has_sig": false,
"md5_digest": "938f8dbe854d231b9d91cd787fb1b1d4",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 7458,
"upload_time": "2024-07-09T10:28:33",
"upload_time_iso_8601": "2024-07-09T10:28:33.677178Z",
"url": "https://files.pythonhosted.org/packages/9d/df/cd933ad4f787bfe46365d9cf4a5ba9db2225744db685335f06741ee608f3/panoptica_genai_protection-0.1.12-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "3319f5ac072ab9ec75a64d8421b48f359bf314977b47f040565154165c97b4f2",
"md5": "c28a3cbcad0cf8c68958fae11a0e6e35",
"sha256": "afc5dd87a2764c4256113ce88ff9df2822d8d3e699c6eaf4e1459739b3999f7d"
},
"downloads": -1,
"filename": "panoptica_genai_protection-0.1.12.tar.gz",
"has_sig": false,
"md5_digest": "c28a3cbcad0cf8c68958fae11a0e6e35",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 3936,
"upload_time": "2024-07-09T10:28:35",
"upload_time_iso_8601": "2024-07-09T10:28:35.263771Z",
"url": "https://files.pythonhosted.org/packages/33/19/f5ac072ab9ec75a64d8421b48f359bf314977b47f040565154165c97b4f2/panoptica_genai_protection-0.1.12.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-07-09 10:28:35",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "panoptica-genai-protection"
}
JSON