Cognigy Command Line Interface (CLI)

Cognigy CLI offers a series of tools and functionalities in order to interact with your Cognigy.AI virtual agent projects from the command line.

The CLI can only create Snapshots and Locales right now - no other resources. It is meant for changing already existing resources and manipulating their state.

Currently supported resources (<resourceType>):

• Flows (clone, restore, push, pull, diff, train)
• Lexicons (clone, restore, push, pull, diff)
• Endpoints (clone, restore, push, pull, diff)
• Snapshots (create)
• Extensions (pull)

For Endpoints, Transformers will be separately stores as TypeScript files

Installation

Install globally

We recommend to install the Cognigy.AI CLI globally to use wherever you like. In order to so, perform the following step.

npm install -g @cognigy/cognigy-cli

Use locally

Alternatively you can use a local copy by performing the following steps:

1. Clone the repo
2. Run npm ci
3. Run npm run build
4. Run node build/cognigy init
5. (optional) Run npm link to enable running cognigy globally

Proxy Configuration

The Cognigy CLI will respect the HTTP_PROXY/http_proxy and HTTPS_PROXY/https_proxy and NO_PROXY/no_proxy environment variables.

If you're behind a proxy, you might need to configure NPM separately to use it. You can do that using the commands below.

npm config set proxy http://usr:pwd@proxy.company.com:8080
npm config set https-proxy http://usr:pwd@proxy.company.com:8080

Configuration

Configuration can be provided either as environment variables or inside a config.json file. The config.json can be created by executing cognigy init.

The Cognigy.AI CLI needs the following configuration:

Key	Description	Environment Var
baseUrl	The base URL of your Cognigy.AI environment API	CAI_BASEURL
apiKey	The Cognigy.AI API Key to use	CAI_APIKEY
agent	The ID of your agent project	CAI_AGENT
agentDir	The folder where you want to store the agent locally	CAI_AGENTDIR
playbookTimeoutSeconds	Timeout for checking playbook runs	CAI_PLAYBOOK_TIMEOUT_SECONDS

Both environment configuration and file-based configuration can be used together. Environment configuration takes priority.

Commands

General Commands

Help

To get help on any command, use the -h flag.

cognigy -h or cognigy <command> -h

Forcing configuration files

By default the CLI will use the ./config.json configuration file. You can force the use of another configuration file by using the -c flag.

cognigy <command> -c ./config2.json or cognigy <command> --config ./config2.json

Forcing execution without warnings

By default the CLI will reconfirm if data is overwritten locally or on Cognigy.AI. You can force these checks to be skipped with the -y flag.

Example: cognigy clone -y or cognigy clone --forceYes

Some environments may have rate limits. This means that when executing commands that require a high number of API requests (e.g. clone), the env rate limit may be reached, and the process stops by throwing an error.

Command: init

cognigy init

Initializes a new Cognigy.AI CLI project

Command: clone

cognigy clone

Clones a Virtual Agent from Cognigy.AI to disk

Option	Alias	Type	Default	Description
--type	-t	String	agent	Which type of resource to clone (agent stands for the full project)

Command: restore

cognigy restore

Restores the local agent copy back into Cognigy.AI by executing a push for every resource.

Option	Alias	Type	Default	Description
--type	-t	String	agent	Which type of resource to restore (agent stands for the full project)

Command: pull

cognigy pull <resourceType> <resourceName>

Pulls a specific remote resource from Cognigy.AI

Command: push

cognigy push <resourceType> <resourceName>

Pushes a specific remote resource to Cognigy.AI

You can't create resources or agents by pushing. Resources must already exist on Cognigy.AI

For Flows, only Intents and Node configurations will be updated, not Flow structure

Command: diff

cognigy diff <resourceType> <resourceName>

Compares a local resource to a remote resource

Option	Alias	Type	Default	Description
--mode	-m	String	full	Full (full) vs Node-by-Node (node) comparison of the Flow

Command: train

cognigy train flow <flowName>

Trains the NLU model of a specified Flow on Cognigy.AI

Option	Alias	Type	Default	Description
--timeout	-t	Number	10000	Timeout in ms before training progress is no longer checked (training will continue on Cognigy.AI)

Command: create

cognigy create snapshot <resourceName> [resourceDescription]

Creates a remote resource on Cognigy.AI and downloads it to disk.

Currently only Snapshots and Locales can be created For Snapshots, use cognigy create snapshot snapname "My Description" For Locales, use cognigy create locale localename

Option	Alias	Type	Default	Description
--timeout	-t	Number	100000	Timeout in ms before the creation process is no longer checked (will continue on Cognigy.AI)
--skipDownload	-s	Boolean	false	Skips download of created resource (for snapshots)
--fallbackLocale	-lf	String	-	ID (not Reference ID) of Locale to use for new Locale
--nluLanguage	-lnlu	String	-	NLU Language to set for new Locale

Command: exportcsv

cognigy exportcsv flow <flowName>

Exports the content of a Flow to CSV.

This command will go through all Flows in all Locales and create a content.csv file next to the JSON. This file can be used to update content.

Command: importcsv

cognigy importcsv flow <flowName>

Imports the content of a CSV back into a Flow.

This command will go through all Flows in all Locales and check if a valid content.csv exists. If yes, it will go through the Flow Chart and update all localized Nodes with the content from the CSV.

Command: localize

cognigy localize flow <flowName>

Adds localization to Flow Intents and Nodes in bulk.

This command will go through all Intents and Nodes in a given Locale and will add a localization if not already present.

Option	Alias	Type	Default	Description
--localName	-l	String	-	Locale to add localizations to
--sourceLocale	-sl	String	-	Source Locale to create localization from (optional)
--sourceLocale	-sl	String	-	Source Locale to create localization from (optional)
--localizeIntents	-li	Boolean	true	Whether to localize intents (if active, localizeNodes default is false)
--localizeNodes	-ln	Boolean	true	Whether to localize Nodes (if active, localizeIntents default is false)
--contentOnly	-co	Boolean	false	Only localize Say, Question and Optional Question Nodes (optional)
--reverse	-r	Boolean	false	Removes the localization from the selected Flow (can be combined with -li and -ln)

Command: translate

cognigy translate <resourceType> <resourceName>

Translates the plain text of a chosen resource, such all Nodes inside a Flow.

Option	Alias	Type	Default	Description
--localeName	-l	String	-	The locale to process
--fromLanguage	-fl	String	-	The language to translate from
--toLanguage	-tl	String	-	The language to translate to
--translator	-tr	String	-	The translation tool that should be used. 'google', 'microsoft' or 'deepl'
--translateIntents	-ti	Boolean	false	Whether to add localization to Flow Intents
--translateNodes	-tn	Boolean	false	Whether to add localization to Flow Nodes
--apiKey	-k	String	-	The API Key for the chosen translation tool
--forceYes	-y	Boolean	false	Whether to skip warnings and overwrite all content

Command: execute

cognigy execute <command>

Executes a command of the Cognigy REST API Client. For more information on API calls, please see our OpenAPI documentation.

Supports injecting payloads either through pipes or the -d (--data) option:

• echo '{"flowId": "5f5618bce35138ed3ab9ab9a"}' | cognigy execute readFlow
• cognigy execute readFlow -d '{"flowId": "5f5618bce35138ed3ab9ab9a"}'

This command uses the baseUrl and apiKey parameters of your configuration.

Option	Alias	Type	Default	Description
--list	-l	-	-	Lists all available commands
--data	-d	string	-	Injects a data payload (must be in JSON format)

Command: run

cognigy run playbooks <playbookFile>

Schedules runs of one or more playbooks and checks their status.

Will either use a ./playbooks.json file or any other file you point it to. Format:

[
    {
        "playbookId": "",
        "entrypoint": "", // snapshot ID or agent ID (if no snapshot is used)
        "flowId": "", // this is the flow reference ID
        "localeId": "" // this is the locale reference ID
    },
    {
        "playbookId": "",
        "entrypoint": "",
        "flowId": "",
        "localeId": ""
    }
]

The run command outputs the status of the playbook runs and exits:

Output	Exit Code
SUCCESS	0
FAILURE	1
TIMEOUT	2

All details are written to ./playbookRunResults.json

Command: knowledge-ai

Cognigy Knowledge AI Documentation

FAQ

Frequently asked questions (FAQ)

Contributing

Make sure you pull origin from the develop branch ```git pull develop``````

Commiting

Commit using the commitizen hook with semantic naming convetion promt

npx cz

Pull Requests

Create PR with any kind of feature/bugfix folloving the semantic message format to the develop branch.

Any PRs to develop needs to be merged as squash merges.

Release

Create a PR from develop to main and do a merge commit. This will automatically trigger a new release. To make the release publish a new minor version to the npm registry, the commit message needs to follow the [semantic message format] and having at least one of the commits to main from the last release with a fix.