Triage the support queue,route it as clean tickets
A support queue is noise: angry paragraphs, one-line “it's broken” reports, duplicates of a bug you already logged. Khint runs your triage rubric against each one, score the severity, strip the noise into a clean repro, and file it to Jira or Linear, all from one keyboard shortcut.
The queue doesn't wait
for you to read carefully
Every report is the same handful of moves: judge how urgent it is, throw out the noise, spot the duplicate, and rewrite the rest so an engineer can pick it up, and you do them dozens of times a day under pressure. That is exactly the dull discipline that erodes when the queue is twenty deep, and a misjudged severity or a vague repro costs an engineer half a day. Khint compresses the triage move with its three pillars, Agents, Capture, and Memory, wired together behind one palette, so triage happens where the report already lives.
One report, through triage
Working a noisy Friday queue down to filed tickets, one report at a time.
Open a session for the triage shift
Before you work the queue, start a Memory session: call it “Friday triage”. Every Action you run is logged to it, and Khint keeps a compact running summary. That summary rides along on each AI Action, so as you move down the queue the model already knows the release you are triaging against and the recurring issues you have already seen this shift. How Memory works →
Write your priority bar down once, as an Action
Your severity rubric lives in everyone's head and nowhere useful. Turn it into a saved Action, “Triage this report”, that returns a one-line summary, a severity call against your bar, and the question that decides it. No prompt yet? Type the description and press Generate; refine it later with Improve. Or clone the Support Desk starter pack from the Agents page for a working triage Action.
Run it on the messy report
Select a customer message: the angry paragraph, the “it's broken again” one-liner, the wall of steps with no actual error, hit Cmd+Shift+K, and run your triage Action. Back comes the verdict: what it is, how urgent under your rubric, and whether it is a duplicate of something you already logged this shift. The noise is gone and the decision is in front of you.
Reshape it into something engineering can act on
For anything worth filing, run a second Action: a structuring prompt that rewrites the customer's words into a clean title, steps to reproduce, and expected vs actual. If the report came in as a screenshot, Capture turns the image into text first. Pair it with an error-to-ticket Action when a stack trace is attached, and the engineering-ready version is pasted right back where you were working.
Route it to the right tracker
With Atlassian connected, the Jira Create issue row files the structured report into your project, one issue per run, so nothing is routed silently. On Linear? The Create issue row does the same there. Chain triage → structure → Create issue into one Workflow and a raw report becomes a filed ticket in a single keystroke. Each filed issue is logged to your Memory session, so the shift keeps its own record. How integrations work →
The rubric is yours to set
Khint applies the priority bar you wrote: this is one example of what a triage Action might return.
There is no fixed severity model buried in Khint: the calls above are just an example. The judgment lives entirely in the prompt you save, so a team that triages on blast radius, a team that triages on paying-tier impact, and a team with a strict SLA matrix each get the review they asked for. Tighten the bar over time with Improve, and clone a starting point from the Support Desk starter pack on the Agents page rather than writing from scratch.
Why not just paste each report into a chatbot?
You can, and for a queue of twenty reports you'll re-state your severity rubric every chat, get a different standard each time depending on how you phrased it, and copy each result back into the tracker by hand. Khint pins the standard: the triage Action is written once and runs identically on every report, Memory carries the shift context into each call automatically, and the Atlassian and Linear integrations file the issue where engineering will actually see it. Because an Action only sees the text you select, plus the session summary you opted into, your customers' reports aren't sitting in a chat history.
Common questions
How does Khint help a support lead triage a queue?
Triage is the same few moves on every report: read it, decide how urgent it is, strip the noise, and turn what is left into something engineering can act on. Khint sits behind one shortcut (Cmd+Shift+K) so you do all of that without leaving the report: select a messy customer message, run a reusable Action, and get back a clean summary, a severity call against your rubric, and a draft repro. It works wherever the report already lives: your helpdesk, your inbox, a Slack thread pasted into a doc.
Does Khint decide severity for me?
It runs the rubric you wrote. You save an AI Action that contains your priority bar (how you weigh blast radius, workaround, paying-tier impact, whatever your team agreed) and Khint applies it to the report text you selected. There is no hidden severity model inside Khint; the standard is the prompt you control, so two support teams with different bars get different calls from the same shortcut. Tighten it over time with Improve.
Is this the same as turning an error message into a bug?
No. Capturing one error off the screen is a single move: see capturing an error message to a ticket for that. Triage is the queue: a dozen vague, emotional, half-duplicate reports where the hard part is deciding what is urgent and which ones are worth engineering's time at all. A triage Action is tuned to judge and reshape an inbound report, not to digitize a stack trace. Khint helps with both, but they are different jobs.
Can Khint file the triaged issue into Jira or Linear?
Yes. Connect Atlassian once with your email, an API token, and your cloud domain, and the palette gains a Jira Create issue row; if your team runs on Linear, the Create issue row does the same there. Run it on the structured report and Khint opens the issue in your project: one issue per run, so nothing is filed silently. Chain your triage and structuring Actions with a Create issue step into a single Workflow and one keystroke takes a raw report all the way to a filed ticket.
What if the customer sent a screenshot instead of text?
Use Capture from the palette. Extract text draws a region around an error toast or console and drops the recognised text on your clipboard; Capture & ask opens a short Vision chat about a screenshot you can't select, so you can pull out the relevant lines. The raw image becomes text you can run a triage Action on, without retyping.
Where do the customer reports go?
An Action only ever sees the text you selected, plus the compacted summary of your active Memory session if you opted in. Memory sessions live in a local SQLite database on your Mac and are never synced to Khint's servers, and your Jira or Linear credentials sit in your OS keychain. Nothing about a customer's report leaves your device silently.
Try it on your next triage shift
Free with 10 AI actions and 5 captures per day. No credit card. Save the triage Action once, run it on every report after.