CHATBOTS

Slack 'Who Owns This Service' Q&A Bot

Answers ad-hoc Slack questions about who owns a service, its runbook link, and SLA tier by querying your service catalog and replying in-thread with a sourced answer.

CategoryChatbots
Enginepaperclip
Difficultyintermediate
Triggerchat
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerEngineer mentions the bot or runs the slash command in SlackSlack
  • ActionExtract the service name and question intent from the messageOpenAI
  • ActionQuery the Postgres service_catalog for matching servicePostgreSQLPostgres
  • LogicIf no confident match, list top candidates and ask user to choose
  • OutputReply in-thread with owner, on-call, runbook, and SLASlack

What it does

Gives engineers an instant, self-serve answer to 'who owns checkout-api?' or 'what's the SLA on the billing worker?' without pinging a channel. A teammate asks the bot in Slack, it looks the service up in your Postgres service catalog, and replies in the same thread with the owning team, on-call contact, runbook URL, and SLA tier.

When to use it

When ownership knowledge lives in someone's head and incidents stall while people hunt for the right team. Best for orgs with a maintained catalog table but no good front door to it.

How it works

  1. 1An engineer @-mentions the bot or uses a slash command in Slack with a service name or fuzzy phrase.
  2. 2The agent extracts the intended service and intent (owner, runbook, or SLA) from the message.
  3. 3It queries the Postgres `service_catalog` table, matching exact and near names.
  4. 4If no confident match is found, it returns the top candidates and asks the user to pick.
  5. 5It posts a formatted reply in-thread with owner, Slack handle, runbook link, and SLA, citing the catalog row's last-updated date.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect SlackChannels, DMs, threads, mentions.
  2. 2
    Connect PostgresAny Postgres URL — query, write, migrate.
  3. 3
    Connect OpenAIModels, embeddings, files.
  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.