CHATBOTS

Slack /docs command: cited answer from ReadMe knowledge base

A Slack slash command that searches your ReadMe docs, answers the question in the channel with inline source links, and quietly logs a suggestion when the matched docs were thin.

CategoryChatbots
Enginesim
Difficultyintermediate
Triggerwebhook
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerSlack /docs slash command receivedSlack
  • ActionSearch ReadMe knowledge base for matching pagesReadMeReadMe
  • ActionCompose grounded, cited answer with OpenAIOpenAI
  • LogicScore retrieval confidence; flag weak coverage
  • OutputPost answer with source links to Slack threadSlack

What it does

Gives your team a `/docs` slash command in Slack. Someone types a question, an LLM searches your ReadMe knowledge base, and the bot replies in-thread with a concise answer plus links to the exact doc pages it used. If the retrieved pages weakly cover the question, it appends a short "this doc could be clearer" note so writers know what to improve.

When to use it

When support, sales, or engineering keep asking the same product questions and the answers already live in ReadMe but nobody finds them. Turns your published docs into a self-serve answer engine without standing up a separate chat UI.

How it works

  1. 1Slack sends the `/docs` slash-command payload to the gateway webhook.
  2. 2The flow pulls the matching pages from ReadMe via search.
  3. 3OpenAI composes a grounded answer using only the retrieved page text, with citations.
  4. 4A logic step scores retrieval confidence; low confidence flags a doc-gap suggestion.
  5. 5The bot posts the answer and source links back to the Slack channel as a threaded reply.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect SlackChannels, DMs, threads, mentions.
  2. 2
    Connect ReadMeAPI docs, changelog, auth.
  3. 3
    Connect OpenAIModels, embeddings, files.
  4. 4
    Connect HTTP webhookTrigger any URL on agent actions.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    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.