Every few hours

Find questions on X and reply helpfully

Real questions about your topic get found and answered, a few at a time.

Runs: Every few hours
Apps: X
AI: Yes, uses the built-in AI allowance
Access: tweet.read tweet.write

Send this to your assistant

Create a ClickRuntime automation: every 2 hours, search X for 'spreadsheet automation help', pick the genuine questions, and email me a helpful reply draft for the best one with a link to post it.

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

Brands that want to be useful on X spend hours searching for people asking questions they could answer. A bookkeeping firm wants to find owners complaining about tax deadlines, a pet-food brand wants posts from worried dog owners. Manual searching is hit-and-miss, replies are often late, and generic replies sound like spam.

Every few hours, recent posts matching your keywords are searched, those already seen are skipped, and ClickRuntime's built-in AI judges how relevant each one is and drafts a helpful reply. Replies can be posted directly, or the drafts can be emailed to you to approve first.

It suits businesses with a clear niche, support teams who want to find questions early and founders who answer in public to build trust. You keep full control of tone and decide whether a person approves each reply.

What you get out of it

Relevant conversations are found for you
Instead of scrolling search results, you receive only the posts the AI judged relevant, so your limited time is spent on conversations where a reply could help. Focus is the difference between an account that grows through useful replies and one that burns out.
Replies are quick and specific
A reply sent within hours of the question, referring to what the person actually said, earns attention and trust that a delayed generic answer does not. Speed shows you care, and people remember brands that help them without being asked twice.
Irrelevant posts are filtered out
The relevance judgement screens out jokes, spam and off-topic matches of your keywords, which keeps you from replying to something embarrassing or pointless. That protects your brand from replying to a joke or an unrelated post with a sales pitch.
No double replies to the same person
Tracking posts already seen stops the automation from replying twice to a single post, which would look careless and annoy the person you are trying to help. Double replies are an easy way to look like a bot, which you want to avoid.
Approval keeps you in control
Choosing the draft-by-email mode means a person reads every reply before it is posted, which is wise for regulated industries or sensitive topics. That is especially important for finance, health or legal topics where a wrong reply has consequences.

How it works

What happens, step by step

Every few hours, ClickRuntime searches X for recent posts that use your keywords.
Posts it has already looked at are skipped.
AI reads each one, decides whether it is a question you can genuinely help with, and writes a helpful reply.
The reply is posted, or, if you prefer, emailed to you as a draft with a Post button.

Before it runs live

Connect X

What you get

RunsEvery few hours

Find questions on X and reply helpfullytoday, 9:00Succeeded

Email · Reply draft for XTweet: "Anyone know how to auto-copy new form rows into a second sheet?" Draft: "An Apps Script onFormSubmit trigger can do it, happy to share a snippet" Post it →

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

Run it at a different time, more or less often, or only on weekdays
Change what the AI looks for, or how it writes the result
Add another app, for example also send it to Slack
Combine it with other X 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.

find_and_reply_to_relevant_tweets.pyexample

from clickruntime import on_trigger, state, notify, http, credentials, ai

QUERY = '"spreadsheet automation" help -is:retweet lang:en'

@on_trigger
def run(event):
    auth = {"Authorization": "Bearer " + credentials.connection("x", "main")}
    params = {"query": QUERY, "max_results": 25}
    if state.get("since_id"):
        params["since_id"] = state.get("since_id")
    data = http.get("https://api.x.com/2/tweets/search/recent", headers=auth, params=params).json()
    tweets = data.get("data", [])
    if not tweets:
        return {"found": 0}
    state.set("since_id", data["meta"]["newest_id"])
    numbered = "\n".join(f"{i}. {t['text'][:280]}" for i, t in enumerate(tweets))
    pick = ai.generate(
        "Which of these tweets is the best genuine question about spreadsheet automation? "
        "Reply with its number, a newline, then a helpful reply under 250 chars. Reply NONE if none are genuine.\n\n" + numbered,
        max_output_tokens=150,
    ).text.strip()
    if pick.startswith("NONE"):
        return {"found": len(tweets), "picked": None}
    num, _, reply = pick.partition("\n")
    tweet = tweets[int(num.strip(". "))]
    link = f"https://x.com/intent/post?in_reply_to={tweet['id']}"
    notify.email_me("Reply draft for X", f"Tweet: {tweet['text']}\n\nDraft: {reply.strip()}\n\nPost it: {link}")
    return {"picked": tweet["id"]}

FAQ

Common questions

Will X treat these replies as spam?

X restricts unsolicited, repetitive or promotional replies, and accounts that do it risk limits or suspension. Keep replies genuinely helpful, avoid links in every post and use the approval mode so each reply is considered.

How are search keywords chosen?

You provide phrases that signal a question you can answer, such as how do I file quarterly tax or best food for puppy allergies. Your AI assistant can add exclusions like retweets, and refine the list as you see the results.

What if I only want to be alerted, not reply?

Then the automation can send you the relevant posts and suggested replies by email without posting anything. You reply by hand where you choose, which is a good starting point while you assess the quality of matches.

What permissions does it need?

tweet.read tweet.write. Nothing else in your accounts is used.

Does it need AI?

Yes. It uses ClickRuntime's built-in AI, so you don't need an OpenAI or Anthropic key. Every plan includes a monthly AI allowance.

How many runs does it use?

One run each time it runs (every few hours). Checks that find nothing to do and stop right away aren't counted.

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.

Every few hours / dailyAI