GitHub Issue Triage Agent
Labels every new GitHub issue, finds the duplicate before a maintainer does, and leaves one reply that tells the contributor exactly what is missing.
Watch it work before it's live
Run a staged conversation — no account needed. The agent handles it for real while a simulated world answers its tool calls; nothing touches real accounts, and nothing is actually sent.
Webhook retries fire twice when the first attempt times out
Openopened by Priya Raghunathan
On v4.2.1 self-hosted, a webhook whose first delivery hits the 10 second timeout is retried, but the receiver gets both the retry and the original once it finally completes. Steps: create a webhook pointing at an endpoint that sleeps 12 seconds, trigger it once, watch the receiver log two deliveries with the same event id about 40 seconds apart. Expected one delivery per event, or at least the same event id deduped. This started for us after upgrading from 4.1.6, where the same endpoint only ever received one call.
Set up in minutes
Using this template drops you into a guided setup. It asks exactly this, nothing else:
Connect GitHub
One sign-in. The agent acts through your account, scoped to what this template uses.
Connect Slack
One sign-in. The agent acts through your account, scoped to what this template uses.
Triage guide
Your label taxonomy and who owns what - the labels you actually use, what each one means, which team or person owns each area, and what counts as critical. The agent only ever labels and routes from this.
Maintainer notes
The repro details you always end up asking for, the parts of the codebase that break most often, and the tone you want used with contributors.
Choose which agent runs it
NoClick's built-in models work out of the box — or bring Claude Code, Codex, and other coding agents on your own subscription.
Watch it handle a test run
A staged conversation against a simulated world — then it’s live.
About this agent
Most issues die of neglect rather than difficulty: nobody labels them, nobody notices it was already reported in March, and the reporter never hears back. This agent runs the second an issue opens, applies your labels, searches for the duplicate, and leaves one comment that either points at the existing thread or asks for the exact repro details you always end up asking for. Maintainers open a triaged queue instead of a pile.
What people use it for
- Route issues to the owning team - Vodafone cut triage on incoming reports to minutes by sending each one straight to the right queue. Same idea here: your label taxonomy decides the owner, and the comment tells the reporter who has it.
- Kill duplicates on arrival - The agent searches open and closed issues before it writes a word, so a report that already exists gets a link to the original thread instead of a second parallel discussion nobody reconciles later.
- Make thin reports fixable - An issue that just says it is broken gets a friendly, specific ask for the version, the steps and the logs you actually need. Contributors hear back in a minute instead of a week, and usually reply with the missing half.
- Wake the team for real fires - Shopify runs issue digests across 110+ teams without paging anyone unnecessarily. Slack fires only for issues that match your critical bar, so the alert channel stays worth reading.
Before you fork
What do I need to connect before this works?
A GitHub credential with issues read and write on the repositories you want triaged, and one Slack channel for critical alerts. Your label taxonomy, owners and the repro details you want asked for go in as variables during setup. There is nothing to install in the repo and no bot account to create.
What does it cost to run?
It runs once per issue opened, and each run is a short read plus one comment, so the cost tracks your issue volume and not your repo size. A repo taking 20 new issues a day is 20 small runs a day. On a busy public repo, watch the first day before leaving it unattended.
Could it embarrass us in public?
It can only add labels that already exist and leave one comment per issue, 2 to 4 sentences, in the tone you set in the maintainer notes. It never closes, edits or reopens anything and never touches code, so a wrong label is a one click fix. Anything outside your triage guide gets a plain hand off to a maintainer rather than an invented answer, and you can rehearse it on the staged issues before pointing it at a real repo.
Run it with your coding agent
More agents like this
Put GitHub Issue Triage Agent to work
Free to start. Guided setup, a test run against staged conversations, and it's live.