Hero AI

The Hero AI connector enables the execution of advanced language models within Swimlane playbooks, enhancing automation with AI-driven natural language processing.

Hero AI is an advanced AI platform that integrates with Swimlane Turbine to provide intelligent automation capabilities within security workflows. By leveraging the 'call_llm' action, users can execute large language models (LLMs) to generate insights, automate responses, and enhance decision-making processes. The integration allows for dynamic interaction with AI models using custom prompts, enabling tailored outputs that can inform security strategies and operations. This connector empowers end-users to harness the power of AI directly within their security playbooks, streamlining complex tasks and providing a competitive edge in threat detection and response.

Prerequisites

To effectively utilize the Hero AI connector within Swimlane, ensure you have the following prerequisites:

Personal Access Token authentication with the following parameters:

URL: Endpoint for Hero AI API services.

PAT: Your unique Personal Access Token for secure access.

Account ID: Identifier for your specific Hero AI account.

Tenant ID: Identifier for your tenant within Hero AI.

PAT Authentication

Personal access token along with Account ID and Tenant ID is required to authenticate Swimlane Hero AI connector.

Capabilities

This Connector provides the following capabilities:

Call LLM

Notes

To Get PAT(Personal Access Token) Follow below steps:

Go to Profile & user settings( top right corner).

Click on Personal access token tab.

Generate the token and use to authenticate.

Swimlane LLM API will be an external API. In order to have better control over the load on this API, this action has been added to used in playbooks.

Additional Notes

Please pass the Tenant ID that is available on your Turbine instance. By default, you can also pass the value of the Tenant ID as 1 but it is preferred to pass the correct value of the Tenant ID to the asset configuration.

Configurations

Swimlane Hero AI PAT Authentication

Authenticates using Personal Access Token.

Configuration Parameters

Parameter	Description	Type	Required
url	A URL to the target host.	string	Required
account_id	Account ID of the logged in User.	string	Required
tenant_id	The Tenant ID.	string	Required
pat	Personal Access Token	string	Required
verify_ssl	Verify SSL certificate	boolean	Optional
http_proxy	A proxy to route requests through.	string	Optional

Actions

Call LLM

Executes a large language model (LLM) within Swimlane playbooks using a provided 'prompt' to enhance automation capabilities.

Input

Argument Name	Type	Required	Description
prompt	string	Optional	Parameter for Call LLM
temperature	number	Optional	The sampling temperature to use. Range is between 0 and 2. Both limits are exclusive (parameter cannot be set equal to 0 or 2).
max_tokens	number	Optional	Range is between 1 and 4096. Both limits are inclusive (parameter can be set to 1 or 4096).

Input Example

{"json_body":{"prompt":"Tell me about swimlane.","temperature":0.1,"max_tokens":1024}}

Output

Parameter	Type	Description
status_code	number	HTTP status code of the response
reason	string	Response reason phrase
request_id	string	Unique identifier
generated_text	string	Output field: generated_text
finish_reason	string	Response reason phrase

Output Example

{"status_code":200,"response_headers":{"Date":"Fri, 01 Mar 2024 06:26:18 GMT","Content-Type":"application/json; charset=utf-8","Transfer-Encoding":"chunked","Connection":"keep-alive","Vary":"Accept-Encoding","Cache-Control":"no-store","Pragma":"no-cache","X-Frame-Options":"sameorigin","Content-Security-Policy":"frame-ancestors self *.swimlane.app","Strict-Transport-Security":"max-age=31536000; includeSubDomains","Content-Encoding":"gzip"},"reason":"OK","json_body":{"request_id":"1bf002a4-6d28-4c...

Response Headers

Header	Description	Example
Cache-Control	Directives for caching mechanisms	no-store
Connection	HTTP response header: Connection	keep-alive
Content-Encoding	HTTP response header: Content-Encoding	gzip
Content-Security-Policy	HTTP response header: Content-Security-Policy	frame-ancestors self *.swimlane.app
Content-Type	The media type of the resource	application/json; charset=utf-8
Date	The date and time at which the message was originated	Fri, 01 Mar 2024 06:26:18 GMT
Pragma	HTTP response header: Pragma	no-cache
Strict-Transport-Security	HTTP response header: Strict-Transport-Security	max-age=31536000; includeSubDomains
Transfer-Encoding	HTTP response header: Transfer-Encoding	chunked
Vary	HTTP response header: Vary	Accept-Encoding
X-Frame-Options	HTTP response header: X-Frame-Options	sameorigin