TICKET MANAGEMENT
Nightly batch: cluster the Zendesk backlog and merge stale duplicates
Runs on a schedule to scan all open Zendesk tickets, cluster near-identical ones across the whole backlog, merge the duplicates into a canonical ticket per cluster.
How it runs
The automated pipeline, trigger to output.
- TriggerNightly schedule fires
- ActionPull all open tickets and embed textZendesk
- ActionCluster embeddings and pick canonical per groupPostgres
- LogicSkip singletons, keep real duplicate groups
- ActionMerge duplicates into canonical ticketZendesk
- OutputEmail backlog cleanup reportGmail
What it does
Sweeps the entire open Zendesk backlog once a night, finds clusters of duplicate tickets that accumulated during the day, merges each cluster down to one canonical ticket, and reports what was consolidated.
When to use it
When real-time merging is too aggressive but the backlog still fills with duplicates. Use this as a safe overnight cleanup that batches decisions and leaves a clear audit trail.
How it works
- 1A nightly schedule triggers the run.
- 2All open tickets are pulled from Zendesk and their text embedded with OpenAI.
- 3Embeddings are clustered in Postgres and a canonical ticket is chosen per cluster by age.
- 4A branch skips clusters of size one and only acts on real duplicate groups.
- 5Each duplicate group is merged into its canonical ticket via the Zendesk API.
- 6A consolidated cleanup report is emailed to the support lead with cluster counts and merged ticket ids.
Set it up
What you configure once, before turning it on.
- 1Connect ZendeskTickets, queues, knowledge base.
- 2Connect PostgresAny Postgres URL — query, write, migrate.
- 3Connect OpenAIModels, embeddings, files.
- 4Connect GmailRead, draft, send, label.
- 5Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
- 6Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
- 7Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.
More Ticket Management workflows
Deduplicate Discord bug reports against existing Linear issues
Before creating anything, searches Linear for issues matching a new Discord bug report; if a duplicate exists it comments and links the report there, otherwise it opens a fresh…
Weekly reopen-by-agent coaching digest
Aggregates each agent's solved-then-reopened tickets for the week, identifies the most common reopen reason per agent, and emails a private coaching digest to the support manager.
Promote a Discord message to a Linear issue via an emoji reaction
When a moderator adds a designated emoji reaction to any Discord message, an LLM converts that message into a structured Linear issue and threads the link back.
Enrich Discord bug reports with Sentry errors before filing in Linear
Takes a Discord bug report, has an LLM pull out likely error signatures, searches Sentry for matching events.
Route Discord bug reports by severity to Linear or PagerDuty
Classifies each Discord bug report by severity using an LLM, then files normal bugs as Linear issues while escalating critical outages to a PagerDuty incident so on-call gets…
Triage Discord bug threads into structured Linear issues with repro checklists
Watches a Discord bug-report channel, uses an LLM to extract a clean title, severity, and step-by-step reproduction checklist from the messy thread.
Run it inside a business
This workflow drops into a full company template. Import the org, and this is one of the playbooks its agents run.
Run this workflow in your colony.
14-day trial. No DevOps. No Sales call. Provisioned in under a minute.