Every 6h

Know when a model you use is updated

When a model or dataset repo you rely on gets a new commit, you hear about it in Slack within hours.

Runs: Every 6h
Apps: Hugging Face
AI: Not needed
Access: read

Send this to your assistant

Create a ClickRuntime automation: every 6 hours, check the Hugging Face repos meta-llama/Llama-3.1-8B-Instruct and mistralai/Mistral-7B-Instruct-v0.3 and post to #models in Slack when either has a new commit.

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

Your product calls a model hosted on Hugging Face, say a fine-tuned embedding model your team pins by name. The repository owner pushes a new revision, and the behavior of your search results shifts a week later. Nobody can say when the weights changed, because nobody was watching the repo, and now debugging means digging through the commit history to guess which update caused the regression.

Every 6 hours, ClickRuntime asks Hugging Face for the repo's current commit hash and last-modified time, compares them with what it saw on the previous check, and posts to Slack when they differ. The message names the repo and gives the new revision and the update time.

It fits ML engineers, data scientists and anyone who depends on a model or dataset they do not control.

What you get out of it

Know the day a dependency changed
When a model you rely on is updated, you hear about it the same day. A behavior shift in production can be matched to a revision date instead of being a mystery that takes days to explain.
Review upstream changes before adopting them
Pinned models are safe only until someone bumps the pin. A notification gives you the chance to read the model card, run your evaluation set and decide on purpose whether to move to the new revision.
Catch licence or card updates early
Owners sometimes change a model's licence, gating or usage terms alongside a new commit. Being told about the update prompts you to check before shipping, not after a legal question arrives.
Stop checking repos by hand
Opening five model pages every few days is a chore that stops happening. A single channel post replaces the habit, so following several repos costs you nothing in attention until something moves.
Reproducibility gets an anchor
Each alert records the commit hash at that moment. When you later need to say exactly which version produced a set of results, the channel history gives you a dated list of revisions to point to.

How it works

What happens, step by step

Every 6 hours, ClickRuntime checks each Hugging Face model you follow, such as Llama 3.1 or Mistral 7B.
It compares the latest version with the one it saw last time.
When a model has a new commit, it posts to #models in Slack with the model name, the short version code, the update time and a link to its commit history.

Before it runs live

Connect Hugging Face

What you get

RunsEvery 6h

Know when a model you use is updatedtoday, 9:00Succeeded

Slack · #modelsNew commit on mistralai/Mistral-7B-Instruct-v0.3 (a1b2c3d, 2026-10-18T09:12:00Z) https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3/commits/main

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 Hugging Face 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.

alert_on_new_commits_releases_of_a_model_or_repo_i_follow.pyexample

from clickruntime import on_trigger, notify, http, state, report_fetched

REPOS = ["meta-llama/Llama-3.1-8B-Instruct", "mistralai/Mistral-7B-Instruct-v0.3"]

@on_trigger
def run(event):
    shas = state.get("shas") or {}
    changed = []
    for repo in REPOS:
        info = http.get(f"https://huggingface.co/api/models/{repo}").json()
        if shas.get(repo) not in (None, info["sha"]):
            changed.append((repo, info))
        shas[repo] = info["sha"]
    report_fetched(len(changed))
    if not changed and state.get("shas"):
        return {"changed": 0}
    for repo, info in changed:
        notify.slack(f"New commit on {repo} ({info['sha'][:7]}, {info['lastModified']})\n"
                     f"https://huggingface.co/{repo}/commits/main", channel="#models")
    state.set("shas", shas)
    return {"changed": len(changed)}

FAQ

Common questions

Can it watch datasets and Spaces as well as models?

Hugging Face exposes the same commit and last-modified information for models, datasets and Spaces, so the same check works for all three. Tell it the repo ids you want watched, for example org/name, and each is compared separately.

What counts as an update, a new commit or a new release?

Any new commit to the repository's main branch changes its commit hash, and that is what triggers the alert. Tag-based releases are not tracked separately, so a repo that only tags without committing would not be detected.

Will it work for private or gated repositories?

It can, as long as the Hugging Face access token you provide has read access to the repo, and you have accepted the terms for gated ones. The token is stored as a secret, and public repos work without needing any special permission.

What permissions does it need?

read. 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 each time it checks. Checks that find nothing new 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.

Weekly