CHATBOTS

Weekly Handbook Reindex for the HR Concierge

On a schedule, pulls the latest pages from the Confluence handbook space, embeds them, and refreshes the concierge's knowledge index so answers always reflect the current policy.

CategoryChatbots
Enginesim
Difficultyintermediate
Triggerschedule
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerWeekly schedule fires
  • ActionList and read changed Confluence pagesConfluenceConfluence
  • LogicDiff against index: adds, edits, deletes
  • ActionChunk and embed changed pagesOpenAI
  • ActionUpsert embeddings into Postgres storePostgreSQLPostgres
  • OutputPost reindex summary to HR ops channelMicrosoft Teams

What it does

Keeps the HR concierge's answers accurate by rebuilding its search index from the live Confluence handbook on a recurring schedule. Changed and new policy pages get re-embedded; deleted pages get purged.

When to use it

Run this when your handbook changes often (open enrollment, new leave policies, comp updates) and you cannot afford the bot answering from stale content. It removes the manual step of re-syncing knowledge after every edit.

How it works

  1. 1A weekly schedule fires the reindex job.
  2. 2The flow lists all pages in the designated Confluence handbook space and reads those modified since the last run.
  3. 3A logic step diffs the page set against the existing index to find adds, edits, and deletes.
  4. 4Changed pages are chunked and embedded into vector representations.
  5. 5The new embeddings are written to the Postgres knowledge store the concierge queries.
  6. 6A summary of pages added, updated, and removed is posted to the HR ops channel.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect ConfluenceSpaces, pages, blueprints.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect PostgresAny Postgres URL — query, write, migrate.
  4. 4
    Connect Microsoft TeamsChannels, chats, files.
  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.