CHATBOTS

Flag Missing Onboarding Docs When the Bot Can't Find an Answer

When the concierge has to escalate because no help-center article covered the user's onboarding question, it logs the documentation gap and files a Linear task to write…

CategoryChatbots
Enginesim
Difficultyadvanced
Triggerevent
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNo-answer escalation in IntercomIntercomIntercom
  • ActionDistill question into doc-gap topic with OpenAIOpenAI
  • ActionUpsert and count gap in PostgresPostgreSQLPostgres
  • LogicRepeated gap with no doc task?
  • ActionFile missing-doc task in LinearLinearLinear
  • OutputSave task reference for dedupPostgreSQLPostgres

What it does

This turns answer failures into a documentation backlog. When the onboarding bot escalates specifically because its knowledge base had no matching content, the unanswered question is captured, grouped with similar ones, and surfaced to the docs team as a concrete Linear task to write the missing guide.

When to use it

Use it when low-confidence handoffs are caused by content gaps rather than edge cases. It tells you exactly which onboarding articles to write next, ranked by how often users ask.

How it works

  1. 1An Intercom event for a no-source-found escalation triggers the flow.
  2. 2OpenAI distills the user's question into a canonical doc-gap topic.
  3. 3The gap is upserted and counted in Postgres against existing topics.
  4. 4A logic branch fires when a gap is asked repeatedly and has no doc task yet.
  5. 5A Linear task is created requesting the missing onboarding article with example questions.
  6. 6The task reference is saved to Postgres to dedupe.

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 PostgresAny Postgres URL — query, write, migrate.
  4. 4
    Connect LinearIssues, projects, cycles, triage.
  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.