Daily Crisp support digest, delivered to Slack each morning

By General Input

Every weekday morning, get a clear summary of yesterday's Crisp chats posted to Slack, leading with what still needs attention today and who handled what.

Integrations

  • Crisp
  • Slack Bot

Type

Agentic Task

Categories

  • Customer Support

Every weekday at 8:30am, generate a support digest covering yesterday's Crisp live chat conversations and post it to our support Slack channel. This runs on a cron schedule with no human in the loop, so it should produce a finished, ready-to-read summary on its own.

First, gather the previous day's conversations from Crisp using List Conversations. Crisp timestamps are Unix epoch in milliseconds, so compute the window for the prior calendar day in our working timezone and scope the list to conversations that were created or last active in that window. Use the operation's search, filter, and ordering options to focus on the right day, and page through the results until you have all of them. For each conversation, capture its state (unresolved, pending, or resolved), which operator it is assigned to, and how long it has been waiting since the last customer message.

Next, for the conversations that matter, especially anything still unresolved or aging, read the thread with Get Messages In Conversation so you understand what the customer actually asked and how it was handled. You do not need to read every resolved chat end to end, but read enough to identify recurring themes and judge sentiment accurately.

Then write a skimmable digest with short, clearly labeled sections. Lead with Needs attention today: conversations that are still unresolved or aging, sorted so the longest-waiting threads come first, each with the customer, how long it has been waiting, the assigned operator, and a one-line summary. Follow with Volume: the total number of conversations for the day and how many are resolved versus still open. Then Recurring themes and complaints: the topics that came up repeatedly, with rough counts. Then Customer sentiment: an overall read of positive, neutral, or negative with a sentence of justification and any standout unhappy customers. Finally Operator workload: how the conversations were split across operators so it is clear who handled what.

Post the finished digest to our support Slack channel (for example #support) using the Slack Bot Send a Message action. Format it for Slack with mrkdwn, using bold section headers and bullet lists, and keep it tight and scannable. Put the Needs attention today section at the very top so the first thing anyone sees is what to jump on. If there were no conversations yesterday, post a short note that it was a quiet day instead of a full digest.

Related prompts

Explore more prompts
A brand asset library your marketing team actually searchesTurn Mailjet email clicks into ranked HubSpot follow-upsClean out the Looker dashboards and Looks nobody opensLiveKit live operations console for room moderationWake up dormant Keap leads with a researched reasonLiveChat coverage board for planning next week's shiftsPhone routing control panel for LiveKit voice agentsLinkedIn Ads budget pacing dashboard for every client accountGive your team Looker numbers without buying more seatsPause marketing emails to escalated customers, then restore them