Sequentum Cloud API Usage

The Sequentum Cloud API is designed for simplicity and ease of use, enabling users to integrate and interact with their agents seamlessly without going through the Sequentum dashboard. The following API calls are available:

Authentication

The API requires authentication which must be configured in the Sequentum Cloud. You must add an API Key to a user account in the Sequentum Cloud, and then use this API Key in all API calls. We recommend you to create a dedicated user account for the API Key.

Follow the below steps to create an API Key. Only administrators can create access tokens:

1. Open the Users page and select Manage API Keys in the context menu for the user you want to give API access.
   

   

2. Click the New API Key button and create a new key.
   

   

3. API keys have a default expiration of one year from the current date. You can modify or extend the expiry date as needed. Additionally, add optional descriptions to each key for improved management and traceability. 
   

   

4. A dialog box will appear for copying the API Key. Carefully copy the key as it will not be displayed again. Once copied, click "OK". Lost API keys cannot be recovered. You must delete the existing key and generate a new one if it is lost.

The API Key must be added to every API request header. The following request shows how the API Key is added to the Authorization header.

List all agents for Organization and the user

GET api/v1/agent/all

This API endpoint grants access to the organization's agent inventory as well as the user's personal space agent inventory. It returns a list of all agents accessible to the authenticated user, along with detailed information for each agent.

Example:

• API Request

• Python API Request

Output:

List agent details

GET  api/v1/agent/{agentId}

This API call returns comprehensive details for a specified agent in the Sequentum Cloud. The response includes the agent's ID, name, space ID, input parameters, and descriptive, etc. The Id parameter serves as the unique identifier for the agent. This endpoint is commonly used to obtain detailed information about individual agents.

• agentId : The ID of the agent of which you want to get details.

Example:

• API Request

• Python API Request

Output:

Start an agent

POST api/v1/agent/{agentId}/start

Starts a new execution of the specified agent using the provided configuration parameters.

Path Parameter

• agentId (Required ) : The ID of the agent that you want to start the run for. This is a required parameter.

Response Body

• inputParameters (optional - default value is an empty list) : A list of name/value pairs that will be passed to the agent as input parameters.

• parallelism (optional - default value is an empty) : The number of sessions you wish to start for the agent run.

• parallelMaxConcurrency (optional - default value is an empty) : The maximum number of concurrent sessions that you want to be started at the same time.

• parallelExport (optional - default value is an “Combined”) : Defines how the agent's output should be exported during parallel execution. The default value is 'Combined'. Supported parameter values are:

  • Combined

  • Separated

• timeout (optional - default value is an empty) : The timeout of a call is the total time available for a request to be processed and answered.

• isRunSynchronously (optional - default value is False) : Determines whether the API waits for the agent run to complete or returns immediately. If set to true, the call will wait until the run has finished before returning.

• proxyPoolId (optional - default value is an empty) : The ID of the proxy pool you want the agent to use for this run.

• isExclusive (optional - default value is an empty) : This is a boolean parameter (true/false) that specifies whether the agent run should be exclusive, ensuring no other agents can run in parallel on the same resources.

• isWaitOnFailure  (optional - default value is an empty) : This boolean parameter (true/false) specifies whether the agent should wait upon failure before attempting a retry or taking further action.

• logLevel  (optional - default value is “info”) : Specifies the level of logging for the run. Options include: Fatal, Error, Warning, Info.

• logMode  (optional - default value is “Text”) : Defines the mode in which logs will be captured during the run. The options could include:  Text, Text and HTML.
  

List all runs for the agent

GET  api/v1/agent/{agentId}/runs

This endpoint provides access to the complete execution history for a given configuration, returning a list of all recorded runs. The results are presented in descending order of execution time, with the most recent records listed first.

• agentId (Required): The ID of the agent of which you want to get the details of all the runs.

Example:

• API Request

• Python API Request

Output:

Stop a run

POST api/v1/agent/{agentId}/run/{runId}/stop

Stops a specific running agent instance. This endpoint provides the capability to stop an agent execution in progress. By specifying the agent's ID (`{id}`) and the corresponding run ID (`{runId}`), administrators can precisely control and terminate individual agent instances as needed.

• agentId (Required): The ID of the agent to stop its run.

• runId (Required): The ID of the run of the agent which you want to stop.

Example:

• API Request

• Python API Request

List all files for a run

GET  api/v1/agent/{agentId}/run/{runId}/files

This endpoint retrieves data files for a given agent run and provides details of the data files generated by that agent run.

• agentId (Required): The ID of the agent to get the files of its particular run.

• runId (Required): The ID of the run of the Agent ID of which file details are to be extracted. 

Example:

• API Request

• Python API Request

Output:

Download a run file

GET  api/v1/agent/{agnetId}/run/{runId}/file/{fileId}/download

Downloads a specific file associated with an agent run. This endpoint allows retrieval of individual files generated during a specific agent execution.

• agentId (Required): The ID of the agent to get the files of its particular run.

• runId (Required): The ID of the run of the Agent ID of which file is to be extracted. 

• fileId (Required): The ID of the File which you want to download.

Example:

• API Request

• Python API Request

Output:

List All Versions of Agent

Get /api/v1/agent/{agentId}/versions

This endpoint retrieves all versions associated with a specific agent. An agent can have multiple versions, allowing for an audit trail of changes and the ability to revert to previous configurations.

The response returns a list of versions, with the active version always being the one with the highest version number—a key detail for identifying the configuration currently in use.

• agentId (Required): The ID of the agent of which you want to get the versions list.
  

Example:

• API Request

• Python API Request

Output:

Restore to a Previous Agent Version

Post /api/v1/agent/{agentId}/version/{versionNumber}/restore

This endpoint restores a specified older version of an agent’s configuration, making it the new active version.
Instead of directly replacing the current active version, the system creates a new version that is an exact copy of the version being restored. This approach maintains a complete audit trail and enables easy rollbacks if needed.

The newly created version will have the highest version number and will automatically become the agent’s active version.

• agentId (Required): The ID of the agent whose version you want to restore.

• versionNumber(Required): The specific version number to be restored.

Example:

• API Request

• Python API Request

Swagger

The Swagger documentation provides a convenient way of testing API calls.

Click here to view the API Calls on Swagger.

Postman

The Postman documentation provides a convenient way of testing API calls.

Click here to view the API Calls on Postman.