CHATBOTS

HR Concierge in Teams: Handbook Q&A with PII Escalation

An employee asks an HR question in Microsoft Teams and a Confluence-grounded bot answers in the thread.

CategoryChatbots
EngineSim + Paperclip
Difficultyintermediate
Triggerevent
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerEmployee @mentions concierge in TeamsMicrosoft Teams
  • LogicClassify: does the question request personal/PII data?
  • ActionIf PII: notify HR queue to take overMicrosoft Teams
  • ActionElse: search Confluence handbook spaceConfluenceConfluence
  • ActionDraft grounded answer with citationsOpenAI
  • OutputReply in the Teams threadMicrosoft Teams

What it does

Gives employees a self-serve HR concierge inside Microsoft Teams. It answers benefits, leave, and policy questions using the company handbook in Confluence, and refuses to handle anything involving personal data, instead escalating to an HR specialist.

When to use it

Deploy when HR is fielding the same handbook questions repeatedly in Teams, but you need a hard guardrail so the bot never reads back or guesses at an individual's PII (salary, SSN, medical leave records).

How it works

  1. 1An employee @mentions the concierge in a Teams channel or DM.
  2. 2A classifier checks whether the question requests personal or sensitive data about a specific person.
  3. 3If it does, the bot posts a short hand-off note and pings the HR queue to take over the conversation.
  4. 4Otherwise, it searches the Confluence handbook space for the most relevant policy pages.
  5. 5An LLM drafts a grounded answer with citations back to the source page.
  6. 6The bot replies in the same Teams thread with the answer and a link to the handbook.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect Microsoft TeamsChannels, chats, files.
  2. 2
    Connect ConfluenceSpaces, pages, blueprints.
  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.