Changelog
Multiple new AI models are avilable in the AI SQL generation API (gpt-4, gpt-3.5-16k, Anthropic Claude)
Various UX Improvements (Home page, AI Query Helper, Schedule, Email)
Your own LogicLoop data can now be a data source. You can to track the success of your workflow and escalate any unresolved issues.
You can now chain rules together in actions. Create an action subscription to trigger a rule and pass your results as parameters.
You can now specify parameters in either UPPER_CASE or lower_case in action templates. Previously we required all parameters to be UPPER_CASE.
You can now associate an entity ID to tickets generated by LogicLoop rules. This is helpful if a single user can generate multiple alerts and you want a way of associating all the alerts to a single user. Entity IDs are searchable so you can search up all tickets that belong to the same entity.
Additionally, you can choose to "Group items with the same entity ID into a single ticket". This means that if the same user triggers multiple alerts, instead of creating a new ticket for each alert, the alerts will all be grouped into one ticket under "Additional Rules Flagged"
You can now re-alert on duplicate items after a period of time. For example, let's say you have an alert that runs every hour, but you only want it to alert your team for the same issue at most once per day. You can turn on the deduplication & include a 1 day time period so that it does not alert multiple times in one day, but will alert again the next day.
Users who don't know SQL can now use LogicLoop's Visual Query Builder to query for data just by choosing a few drop down filters. If your database is Postgres, MySQL, Redshift, Snowflake, or BigQuery, the option to use the Visual Query Builder will automatically show up in the top right corner of your query's page.
When you sign up for LogicLoop, we now pre-load your account with a sample database and a few rules to help you get started quickly.
LogicLoop now provides autocomplete suggestions for template parameters when editing action or ticket action subscriptions. The parameter options will automatically appear below once you type the opening curly braces.
You can now view the edit history of a query. While viewing a rule, click on the three dots menu button at the top right corner, and select 'View Edit History'.
There is now a new home screen with step by step instructions to help you get started with LogicLoop. In addition, check out our new demo and templates here.
You can now write to DynamoDB as an action. That means you can have your rule insert or update database rows each time it runs.
You can now take actions on your tickets! For workflows that involve a human in the loop, you may want to first schedule a LogicLoop rule that creates a ticket whenever the rule is triggered. Afterwards, a human can follow up on that ticket and manually execute an action on it e.g. Approve or Deny a user.
To create a Ticket Action, first go to the 'Queues' tab on the left navbar and click on 'Settings' to configure your Ticket Actions. You will be able to select from the same actions you've set up under Action Destinations.
Next, you can click on any of your tickets and click 'Take Action' to execute the action on that ticket.
You can now write to Google Sheets as an action. First, create a Google Spreadsheets Action Destination.
Next, attach an Action Subscription to your rule that will write to a specific Google spreadsheet. All you have to do is paste in the URL of your Google sheet and specify which tab number you wish to write to. When you rule runs, it will write all the rows returned from your SQL query into the spreadsheet.
You can now write a query that refers to a list of items returned from another query. In this example below, we retrieve all api_calls whose id belongs to a list of whitelisted ids defined by another query.
To make this happen, just insert a parameter into your query using curly brackets like
{{ param }} and then in the parameter's settings choose 'Query Based Dropdown List', select the query you want to pull from, and choose to automatically select all values.
You can now choose to receive an email notification if there has been an error running a query you wrote. To enable this feature, an admin must first go to Settings > General and turn on email query owners when scheduled queries fail.
.png?alt=media&token=742120ba-0dfd-483a-8676-8568734658c6)
Example of what a failed query email looks like
You can now set up your LogicLoop queries to trigger PagerDuty alerts. Simply create a PagerDuty action destination type, enter your PagerDuty service integration key, and set up your query to trigger a PagerDuty action.
A PagerDuty action
What the alert looks like in PagerDuty
You can now set a rate limit on your action to prevent it from triggering downstream actions too quickly. This comes in handy if your rule returns a large number of rows and your action triggers something like a Twilio or Slack API that has a rate limit.
We've introduced a more user friendly Webhook Builder action destination type to allow you to construct a webhook request. In addition, a technical user on your team can set up a webhook endpoint once, which your non-technical user can then plug and play where ever necessary. More info here.
.png?alt=media&token=db9ccf5e-c660-4248-ab77-e9eb0f244331)
You can now take action on up to 1000 rows per rule run. Previously, we capped the maximum number of rows you can action upon at 500.
You can now include a CSV file of your query results in email destinations.
For API JSON based sources, you can now enter a custom Authorization header when you're configuring the source:
In this update, you can now dedup your queries so that actions are not taken on rows whose
id field has been seen before. All you have to do is turn on the toggle that says Deduplicate under your action settings:In order to enable this feature, your SQL query must return an
idfield. A clever way of doing so is by naming whichever column you wish to designate as the identifier as id in your SQL query e.g.select user_id as id from accounts
In this update, you can now see a history of all actions that your rule has taken. Just go to your action and under history, you can click View Action Logs to see all the actions that were associated with each rule run, as well as its payload and any errors. In order to enable this feature, go to Settings > General and turn on Record and show history of rule runs under Logs.
In this update, you can now see a history of every single time your rule ran. Just go to your action and in the history section you will see a log of every rule run with information on: when it ran, what its status was, how it was triggered, and what conditions it ran under.
In order to enable this feature, go to Settings > General and turn on Record and show history of rule runs under Logs.
In this update, you can now write to your Postgres or MySQL database an action. That means you can insert, update, or delete database rows as a result of your rule run.
You can also now use Firebase as an action destination. If you want the result of your rule run to send your customers push notifications, all you have to do is enter your Firebase credentials and hook it up as an action destination!
In this update, you can now specify the To: field when you configure an email subscription. Previously, you could only hardcode the To: field once when you create the email destination, but now you can configure emails to be sent to different recipients (e.g. each of your customers) by embedding their email in the To: field.
In this release, we introduce LogicLoop’s case management system. You can create, assign, filter, and triage tickets under the Queues tab. You can also have your rule runs automatically generate tickets for triage. You can watch a demo here and read more detailed documentation about the full feature here.
In this release, we’re letting you take more custom and powerful actions on the results of your LogicLoop queries. You can watch a demo here.
Create actions for each row returned
Instead of triggering one action for the entire result of your rule run, you may want to trigger a separate action for each row. For instance, you may want one Slack message per customer that needs attention, or you may want to call an API endpoint for each new customer returned. Now, you can trigger actions for each row in your Trigger preferences.
Customize templates for each action
You may want to send different content in your Slack message and an email alert on the same query. Now, you can customize data sent to each downstream action. You can embed query specific data into your notifications by injecting special variables in curly braces.
Advanced webhook support
Invoke more powerful downstream APIs with advanced webhook configurations. Set authorization headers, method type, individual URLs per row returned, headers and JSON payloads.
Last modified 1mo ago