CONTENT CREATION

On-demand draft fact-check from a Slack message

Paste a draft or a doc link into a Slack channel and get back a threaded reply that splits every claim into sourced, unsourced, and contradicted, with citation links for each.

CategoryContent Creation
Enginepaperclip
Difficultybeginner
Triggerchat
Steps5
Setup~5 min

How it runs

The automated pipeline, trigger to output.

  • TriggerMessage posted in the fact-check Slack channelSlack
  • LogicDetect whether the message is a link or pasted draft text
  • ActionSearch Exa for a source per extracted claimExa
  • LogicClassify claims as sourced, unsourced, or contradicted
  • OutputReply in-thread with the bucketed claims and citationsSlack

What it does

This workflow turns Slack into an instant fact-checking desk. A writer drops draft text (or a link) into a designated channel, and within moments an agent replies in-thread with a breakdown of every factual claim: which ones are already supported, which have no source, and which the web actively contradicts, each with links.

When to use it

Use it for ad hoc checks before a draft ever reaches the formal calendar, or when a writer wants a second opinion on a specific paragraph. It fits teams that live in Slack and want answers without opening another tool.

How it works

  1. 1A message posted in the fact-check Slack channel triggers the run.
  2. 2If the message contains a URL, the workflow fetches the page content; otherwise it uses the pasted text directly.
  3. 3The agent extracts each factual claim and runs an Exa search for a primary source per claim.
  4. 4A classification step sorts claims into sourced, unsourced, and contradicted buckets.
  5. 5The workflow replies in the original Slack thread with the three buckets and citation links for each verified claim.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect SlackChannels, DMs, threads, mentions.
  2. 2
    Connect ExaNeural search across the web.
  3. 3
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  4. 4
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  5. 5
    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.