The moment it happens

Sort incoming messages and flag bugs

Every message gets a label, and you hear about bug reports by email straight away.

Runs: The moment it happens
Apps: Cohere
AI: Not needed
Access: API key

Send this to your assistant

Create a ClickRuntime automation: when text is posted to my link, label it bug, billing or other with Cohere, and email me the full message whenever it's a bug.

Paste it into your assistant. It tells you what it will send or change before it runs live.

Add to ChatGPT Add to Claude

First time? Adding ClickRuntime to your assistant takes about two minutes.

Why use it

Why teams use this automation

Support inboxes and feedback forms are mixed bags. One message reports a crash, another asks about an invoice, a third is a feature idea, and a human has to read each to decide who should see it. When a bug report sits behind twenty billing questions, the one thing that needed attention waits the longest.

Whenever a message is posted to this automation, it sends the text to Cohere's classifier with a set of labeled examples, receives a label such as bug, billing or other, and emails you straight away if the label is bug. Everything else is tagged and left alone.

It is useful for small teams without a ticketing system, and as a first filter in front of one. Your own examples define each label, so the categories fit your product and the way your customers write, and not a generic list of topics.

What you get out of it

Surface bugs ahead of the rest
A message classified as a bug reaches you in minutes. Those are often the ones where a customer is blocked, and the first response matters most. Bug reports are often from customers who are blocked, and an alert within minutes gives support a chance to reply while it matters most.
Stop reading every message to triage
The label is assigned automatically, so a person only gets involved where it matters. A few hundred messages a week stops feeling like a full-time reading job. With the sorting handled for you, a person only reads the messages that need action, and a few hundred a week stops being a job.
Keep the categories your own
The examples define what counts as a bug or billing question in your business. Adding a few more examples improves accuracy for your particular wording and products. Adding a few more examples of what you call a bug or a billing question improves accuracy for your own products and wording.
Handle volume spikes calmly
After a release, bug reports can surge. Each one is classified and alerted without anyone needing to sort them first, so triage keeps pace with the incoming load. A surge of bug reports after a release is classified and alerted as it comes in, so triage keeps pace with the load.
Add structure to messy feedback
Labeled messages can be counted later. That turns scattered comments into a count of bugs against billing questions, which helps guide what to fix first. Counting labels later turns scattered comments into a clear ratio of bugs to billing questions, which helps you choose what to fix.

How it works

What happens, step by step

When a message is sent to your personal ClickRuntime link, the automation starts straight away.
Cohere compares it with a handful of example messages and labels it as a bug, billing or other.
If it is a bug report, you get an email with the full message straight away. Other labels are recorded but not emailed.

Before it runs live

Connect Cohere
Paste your automation's link into the other app's webhook settings

What you get

RunsThe moment it happens

Sort incoming messages and flag bugsjust nowSucceeded

Email · Bug reportsNew bug report: Export to CSV fails with a timeout for workspaces over 10k rows.

Build your own

Need it to work differently?

This template is a starting point. Copy the prompt and change it, or describe a different automation from scratch, with your own conditions, more apps or an AI step. Your assistant writes it, tests it and runs it on ClickRuntime.

Your own rules
Conditions, thresholds, filters and exceptions, written the way you'd explain them to a colleague.
Any number of apps
One automation can read from one app and update several others. Over 248 apps are ready, and anything with an API works too.
AI where it helps
Summaries, classification, extraction and drafts use ClickRuntime's built-in AI, with no extra key.
On a schedule or on each event
Runs as often as every 5 minutes, or the moment an app sends an event, and keeps track of what it already handled.

Changes you can ask for

Act only on the items that match your own rules, and skip the rest
Add an AI step, for example to summarize, sort or draft a reply
Add another app, for example also send it to Slack
Combine it with other Cohere steps into one automation

Under the hood

The code that runs this automation

This is the Python script that runs each time the automation starts. Your AI assistant writes it for you when you send the prompt, and you allow each step in your assistant. To change how the automation works, describe the change to your assistant and it updates the script for you. For every function it can use, see the script reference.

classify_posted_items_into_labels.pyexample

from clickruntime import on_event, credentials, notify, http

EXAMPLES = [
    {"text": "The app crashes when I upload a file", "label": "bug"},
    {"text": "Error 500 on the settings page", "label": "bug"},
    {"text": "I was charged twice this month", "label": "billing"},
    {"text": "How do I update my credit card?", "label": "billing"},
    {"text": "Do you have a partner program?", "label": "other"},
    {"text": "Love the product, thanks!", "label": "other"},
]

@on_event
def handle(event):
    text = (event.json or {}).get("text") or event.body
    resp = http.post("https://api.cohere.com/v1/classify",
                     headers={"Authorization": f"Bearer {credentials.secret('COHERE_API_KEY')}"},
                     json={"inputs": [text], "examples": EXAMPLES})
    resp.raise_for_status()
    label = resp.json()["classifications"][0]["prediction"]
    if label == "bug":
        notify.email_me("New bug report", text)
    return {"label": label}

FAQ

Common questions

How does the classification work?

The automation sends the message to Cohere's classify endpoint along with example messages for each label. Cohere returns the label it thinks fits best, which is then used to decide whether to email you. The message and your labeled examples are sent to Cohere's classify endpoint, which returns its best label, and that decides whether an email goes out.

Can I add more labels?

Yes. Add a few examples for each new label, such as refund request or feature idea, and tell your AI assistant to include them. The updated examples are shown for your approval first. A new label such as refund request needs a few examples of its own.

What happens with a message that fits nothing?

It receives the closest label, typically other, and no alert is sent. If many messages land there wrongly, adding more examples of the missed kind usually improves the result. The closest label is applied to anything that fits nothing, usually other, and more examples of the missed kind will improve later results.

What permissions does it need?

API key. Nothing else in your accounts is used.

Does it need AI?

No. It runs without AI, so it doesn't use your plan's AI allowance.

How many runs does it use?

One run for each event the app sends.

Can I change how it works?

Yes. Describe the change to your AI assistant, for example a different schedule, channel or filter. It updates the automation, tests it and tells you what changed.

The moment it happens