CHATBOTS

Discord Bot: Is This Test Flaky or Really Broken?

A Discord command that takes a failing test name, queries historical CI run data in Axiom.

CategoryChatbots
Enginesim
Difficultyadvanced
Triggerchat
Steps4
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerDiscord slash command with test name and branchDiscordDiscord
  • ActionQuery Axiom for test outcome historyAxiom
  • LogicCompute flake rate and classify flaky vs regression
  • OutputReply with verdict, flake rate, and recommendation in DiscordDiscordDiscord

What it does

When a contributor's pipeline fails on a test, they ask the bot whether it's their fault. The bot queries Axiom for that test's recent pass/fail history across all pipelines and answers whether the failure looks like a known intermittent flake or a real regression that their change likely introduced.

When to use it

Use this on repos plagued by flaky tests, where contributors waste time debugging failures that aren't theirs. It gives a data-backed verdict ("this test failed 14 of the last 200 runs on unrelated branches — likely flaky") instead of a guess.

How it works

  1. 1A contributor runs a Discord slash command with the failing test name and their branch.
  2. 2The bot queries Axiom for the test's outcome history over a recent window.
  3. 3A logic step computes the baseline flake rate and compares it against this branch's behavior.
  4. 4It classifies the result as likely-flaky, likely-regression, or inconclusive.
  5. 5The bot replies in Discord with the verdict, the flake rate, and a recommendation to retry or investigate.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect DiscordCommunity channels + voice + bots.
  2. 2
    Connect AxiomLog streams, queries, dashboards.
  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.