CHATBOTS

Voice-answer Intercom FAQ grounded in your Help Center

Answers accessibility-mode Intercom questions by retrieving matching Help Center articles, summarizing them with OpenAI.

CategoryChatbots
Enginesim
Difficultyintermediate
Triggerwebhook
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerIntercom accessibility-mode question webhookIntercomIntercom
  • ActionSearch Help Center articlesIntercomIntercom
  • LogicBranch: route to human if no confident match
  • ActionSummarize top article with OpenAIOpenAI
  • ActionRender spoken answer with ElevenLabsElevenLabsElevenLabs
  • OutputReply in Intercom with audio + source linkIntercomIntercom

What it does

This turns your existing Help Center into spoken answers. When an accessibility-mode visitor asks a question in Intercom, the workflow searches your published articles, summarizes the best match into a short answer, and replies with audio so the visitor can listen to the solution.

When to use it

Use it when most inbound questions are repetitive how-to or policy questions already documented in Intercom's Help Center, and you want grounded, source-backed voice answers rather than free-form LLM guesses.

How it works

  1. 1An Intercom webhook fires when a flagged accessibility-mode visitor sends a question.
  2. 2An action queries the Intercom Help Center search API for articles matching the message text.
  3. 3A logic step checks confidence: if no article scores high enough, it routes to a human teammate instead of guessing.
  4. 4OpenAI summarizes the top article into a spoken-friendly answer, keeping it short and free of markup.
  5. 5ElevenLabs renders the summary to speech.
  6. 6The output step posts the audio reply with the linked source article for sighted teammates and auditing.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect IntercomConversations, contacts, articles.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect ElevenLabsText-to-speech, voice cloning.
  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.