Docker Hub Image Intelligence — Pulls, Stars & Health (AI)

Before you base a container on a Docker Hub image, know what you're pulling in. Give this Actor a list of Docker Hub image references and it turns each one into a clear intelligence record: how many times it's been pulled, how many stars it has, how many tags it publishes, how big it is, when it was last updated, whether it's an official image — and an AI risk read that tells you plainly whether an image is active, stale, or effectively abandoned.

You bring the image names. It hands back a container due-diligence sheet.

---

The flow

1. Paste Docker Hub images         2. The Actor gathers facts            3. Get an AI risk read
   (nginx, redis, postgres,   →       Docker Hub registry API       →      pulls · stars · tags · size ·
    grafana/grafana, ...)              (public, no API key)                 update recency · flags

1. Paste Docker Hub image references into the images field — one per line. Official images are just a name (nginx, redis); namespaced images use owner/repo (grafana/grafana).
2. The Actor pulls public data for each: the Docker Hub registry API for repo metadata (pulls, stars, last updated, official status) and the tags endpoint for tag count, latest tag, image size and release cadence. No API key, no scraping of login-walled pages.
3. The AI layer reads the facts and returns a maintenance signal, a popularity tier, concrete risk flags and a two-sentence summary — grounded only in the data, never invented.

---

What you get per image

For every image, the Actor returns:

1. Identity & status — image, namespace, repo, isOfficial, status, isAutomated and description.
2. Reach — totalPulls (lifetime pull count) and stars, so you can tell a ubiquitous base image from a weekend experiment.
3. Footprint — tagCount, latestTag and compressedSizeBytes (compressed download size of the latest tag).
4. Maintenance signals — lastUpdated (ISO timestamp), lastUpdatedAgeDays (how long since the newest push) and releaseSpanDays (how far apart the published tags span — a cadence hint).

Then the AI layer (OpenAI gpt-4o-mini, a fast and low-cost model) adds:

• maintenanceSignal — active, maintained, stale, or abandoned, from update recency.
• popularityTier — niche, growing, popular, or ubiquitous, from pull volume.
• riskFlags — short, data-backed phrases: "unmaintained (not updated >18mo)", "not an official image", "single tag only", "very few tags", "large image size", "low adoption".
• aiSummary — two factual sentences on the image and its risk posture.
• suggestedAlternatives — only well-known, widely-trusted substitutes for the same job, else [].

---

Three ways people use it

• Container image due diligence — before a new image lands in a Dockerfile or compose.yaml, run a Docker Hub image analysis to see pulls, stars, size and whether it's still maintained. Catch an unofficial image, a single-tag repo or a stale base before it's in your build.
• Supply-chain risk — audit an existing image list for container supply-chain risk: surface unmaintained images, unofficial repos and bloated sizes across your whole stack in one run. A standing input for your supply chain security reviews.
• Image selection — comparing candidates for the same job? Line them up and let pull tier, maintenance signal and suggestedAlternatives make the call obvious.

Keywords: Docker Hub image analysis, container supply chain, unmaintained images, image size.

---

Input

Field	Type	Notes
images	array	Required. Docker Hub image references, one per line (e.g. nginx, grafana/grafana). Tags, digests and hub.docker.com/... URLs are tolerated and normalised.
maxResults	integer	Cap how many images to analyze. Default 200.
maxConcurrency	integer	Images processed in parallel. Default 6.
model	string	OpenAI model override. Default gpt-4o-mini.

Minimal input example

{
  "images": ["nginx", "redis", "postgres"]
}

---

Output example

{
  "image": "nginx",
  "namespace": "library",
  "repo": "nginx",
  "isOfficial": true,
  "description": "Official build of Nginx.",
  "totalPulls": 13458027388,
  "stars": 21408,
  "lastUpdated": "2026-10-08T13:52:15.035620Z",
  "lastUpdatedAgeDays": 4,
  "tagCount": 1339,
  "latestTag": "stable-trixie-perl",
  "compressedSizeBytes": 75775660,
  "releaseSpanDays": 12,
  "isAutomated": false,
  "status": "active",
  "maintenanceSignal": "active",
  "popularityTier": "ubiquitous",
  "riskFlags": [],
  "suggestedAlternatives": ["httpd", "caddy"],
  "aiSummary": "nginx is a ubiquitous official image with over 13 billion lifetime pulls and more than 1,300 published tags. It was updated within the last week, showing active maintenance and no significant supply-chain risk."
}

An image that is not found on Docker Hub comes back with the same shape plus an error field (and is not charged).

---

Pay-per-event pricing

This Actor uses pay-per-event billing with a single event:

Event	Price	Charged when
image-analyzed	$0.10	Once per image successfully analyzed (facts gathered + AI risk read produced).

The Actor logs its OpenAI token usage and the estimated AI cost per image at the end of every run, so the event price always stays safely above the model cost.

---

Data source

Everything comes from public, key-less sources: the Docker Hub registry API — https://hub.docker.com/v2/repositories/<namespace>/<repo>/ for repo metadata (pulls, stars, last updated, official status) and https://hub.docker.com/v2/repositories/<namespace>/<repo>/tags/ for tag count, latest tag, image size and release cadence. Official images use the library namespace, resolved automatically from a bare name. Rate limiting or a missing tags page is handled gracefully — the record still ships, just without tag details.

---

Setup

The AI layer calls the OpenAI API. Set the environment variable / secret OPENAI_API_KEY on the Actor. The fact-gathering step (pulls, stars, tags, size, update recency) needs no key.

---

Use it from an AI agent (Apify MCP server)

This Actor is available through the Apify MCP server, so AI agents (Claude, ChatGPT, or anything that speaks MCP) can call it as a tool — hand it a list of images and read back the intelligence records — with no custom integration code.

---

Run it locally

npm install
# put your input in storage/key_value_stores/default/INPUT.json
export OPENAI_API_KEY=sk-...
apify run -p

---

Related actors

• npm Package Intelligence — downloads, dependencies, maintenance and AI risk for any npm package.
• PyPI Package Intelligence — the Python twin: downloads, dependencies, maintenance and AI risk for any PyPI package.
• GitHub Repo Analyzer — health, activity and maintenance signals for any GitHub repository.
• CVE Vulnerability Lookup — look up known vulnerabilities and severity for packages and software.

---

Part of the No Niner toolkit — no-code web scraping, enrichment & AI data tools, all pay-per-use. Browse the full set at noniner.io.