Issue Summary

Slack enforces the limit of 1 message per second which can result in Atlassian Assist being rate-limited.

The specific rate limiting that applies to Assist is Posting Messages. 

From https://api.slack.com/apis/rate-limits:

Posting messages

1 per second

Short bursts >1 allowed. If you attempt bursts, there is no guarantee that messages will be stored or displayed to users. If the burst exceeds available limits, users will see an error message indicating that some messages from your app are not being displayed.

Burst limiting 

You'll see mentions of burst tolerance in the chart above; burst limits are similar to rate limits. While a rate limit defines the maximum requests allowed in a specific timeframe (typically per minute), a burst limit defines the maximum rate of requests allowed concurrently.

Slack does not share precise burst limits externally, because these numbers are subject to change and we don't want you to to build your app with a specific burst capacity in mind only to have to change it later. However, we do recommend you design your apps with a limit of 1 request per second for any given API call, knowing that we'll allow it to go over this limit as long as this is only a temporary burst.

"Why even have burst limits?" you might ask. Slack is primarily a communication tool for humans. We try to detect apps acting spammy, unintentionally or not, and quiet them down to avoid hindering users' ability to communicate and use their workspace's archive.

Steps to Reproduce

1. Have a large number of requests connected to Slack via Assist. 
2. Prepare an automation that will modify those(add a comment or close). 
3. If updates are greater than 1 request per second, you can be rate-limited. 

Expected Results

Atlassian Assist will not be rate limited. 

Actual Results

Atlassian Assist will display a rate-limiting message from Slack:

Workaround

Frequently automation can trigger the rate-limiting if a large number of issues are commented on or change status. You can modify your automation to reduce the number of issues affected.