SOCIAL MEDIA

Agent-Driven Carousel Brief to A/B Variants with Discord Recap

A chat brief drives an agent that writes two creative directions, renders each as a carousel on Replicate, schedules both.

CategorySocial Media
Enginepaperclip
Difficultyadvanced
Triggerchat
Steps7
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerChat message with campaign brief
  • ActionAgent drafts two creative directions
  • ActionRender a carousel per direction on ReplicateReplicateReplicate
  • ActionSchedule both variants across platformsSocial publishing
  • ActionPull engagement per variant after windowSocial publishing
  • LogicCrown winner by engagement rate
  • OutputPost winner recap to DiscordDiscordDiscord

What it does

Turns a plain-language chat brief into a full A/B carousel experiment. An agent interprets the brief, drafts two distinct creative directions, renders each on Replicate, schedules both across platforms, and after measurement crowns a winner and drops a readable recap into Discord.

When to use it

Use this when you want to brief a campaign conversationally and have the agent handle creative variation, scheduling, and reporting. Best for small teams that live in chat and want experiments without a rigid intake form.

How it works

  1. 1A chat message with a campaign brief triggers the agent.
  2. 2The agent interprets the brief and writes two distinct creative directions.
  3. 3Replicate renders a carousel for each direction.
  4. 4post-to-platforms schedules both variants with tracking tags.
  5. 5After the measurement window, the agent pulls engagement per variant.
  6. 6A logic step crowns the winner by engagement rate.
  7. 7A formatted recap with winner, metrics, and next-step ideas posts to Discord.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect ReplicateImage, video, and model inference.
  2. 2
    Connect Social publishingCross-post to X, LinkedIn, Instagram, TikTok, and 4 more in one call.
  3. 3
    Connect DiscordCommunity channels + voice + bots.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.