Skip to content
Get packs

Guide · Agent skill

How to use the Content Analytics via Apify agent skill

Updated · written with AI from the project's own documentation and checked against our catalogue data

The apify-content-analytics skill gives an AI agent a repeatable way to collect public engagement data from Instagram, Facebook, YouTube and TikTok, using third-party scrapers hosted on Apify, a paid web-scraping service. It is written for marketers, creators and analysts who want to compare post and video performance. After reading this guide you will know what the skill does, how to install it, how to ask for it in plain language, and what you need to set up before a paid run.

Content Analytics via Apifysickn33/agentic-awesome-skills · MIT

Quick start

  1. 1Install the skill. Download the ZIP from this page and unzip it into ~/.claude/skills/, or into a project's .claude/skills/ folder. Alternatively, upload the ZIP under Settings → Capabilities → Skills in the Claude apps.
  2. 2Create a local .env file. Add APIFY_TOKEN=your_token to a .env file in your working folder. Keep the file private and out of version control.
  3. 3Check Node.js. Make sure Node.js 20.6 or newer is installed, since the bundled script uses native --env-file support.
  4. 4Make a natural request. Describe the platform, content type and volume you want analysed, for example recent Instagram reels or TikTok posts for a hashtag.
  5. 5Confirm the Actor and limits. The agent selects a scraper, inspects its input schema, and asks you to confirm the Actor, output format, file name and result count, including the paid cost.
  6. 6Review the export and summary. Open the CSV or JSON file, or read the quick answer in chat, then check the summary's item count, file location and insights.

What is an agent skill and what does this one do?

An agent skill is a folder containing a SKILL.md instruction file and optional scripts that an AI agent can load when a task matches the skill's description. This one is named apify-content-analytics and is described as tracking engagement metrics and analysing content performance across Instagram, Facebook, YouTube and TikTok.

Once installed, the agent can collect public engagement data such as likes, comments, views, post and reel performance, hashtags, ads and follower counts. It picks a suitable Apify Actor (a hosted scraper) for the platform and data type, asks about output format and size, runs a bundled Node.js script that calls Apify's API, and then summarises which content performed best.

  • Output as a quick answer in chat (up to five records)
  • Output as a CSV file
  • Output as a JSON file
  • A summary including number of items analysed, file location, key insights and suggested next steps

How do you install the Content Analytics skill?

You can install the skill in one of two ways. The first is to download the ZIP from this page and unzip it into your personal skills folder, ~/.claude/skills/, or into a project's .claude/skills/ folder if you want it available only for that project.

The second is to upload the ZIP in the Claude apps under Settings → Capabilities → Skills. Use whichever option matches how you work with Claude.

  • Personal install: unzip into ~/.claude/skills/
  • Project install: unzip into the project's .claude/skills/ folder
  • Claude apps: Settings → Capabilities → Skills, then upload the ZIP

What do you need before running it?

Check these requirements before any paid run. The skill's instructions say to verify prerequisites first, because each run can incur charges.

Your Apify token must be stored locally in a .env file as APIFY_TOKEN. Do not paste the token into chat or into command arguments. Keep the .env file untracked and private.

  • Node.js 20.6 or newer (needed for native --env-file support)
  • An Apify account with paid credits and an APIFY_TOKEN in a local .env file
  • Apify connector access or Apify Console access, so the agent can inspect Actor schemas
  • Network access to Apify's API and to the chosen platform scrapers
APIFY_TOKEN=your_token

How do you trigger the skill with a natural request?

You do not need special syntax. Describe the content analysis you want in plain language, including the platform, the content type and roughly how much data you need. The agent should then ask about output format and the number of results before running anything.

Before launch the skill asks you to confirm the selected Actor, the public-data scope, the input schema, result limits and pricing. Start with a small input limit.

  • "Compare my last 50 Instagram reels by engagement and tell me which performed best."
  • "Export the top TikTok posts for #yourbrand as a CSV."
  • "Show me the five most-viewed YouTube Shorts from this channel in chat."
  • "Check which Facebook ads from this page had the most reactions, as a JSON file."

What does the script run and what does it touch?

The skill runs a bundled script, reference/scripts/run_actor.js, which calls Apify's API using your local token. The script sends a skill-identifying User-Agent to Apify. --max-items limits only the downloaded output (default 1000) and does not cap the Actor's work or billing, so check Actor settings and pricing yourself.

The skill tells the agent to collect only authorised public data. It does not permit collecting private group data or unnecessary personal contact information, and it advises minimising or redacting personal fields before sharing results. Respect each platform's terms and access controls.

  • Runs scripts: yes (reference/scripts/run_actor.js)
  • Uses network: yes (Apify's API and platform scrapers)
  • If a run times out locally, the remote run keeps going; check the printed run ID in Apify Console before retrying to avoid duplicate charges
  • Output files must be new paths; the agent will not overwrite existing files
node --env-file=.env "<resolved-skill-directory>/reference/scripts/run_actor.js" --actor "ACTOR_ID" --input 'JSON_INPUT' --output YYYY-MM-DD_OUTPUT_FILE.csv --format csv

How should you read the results without overclaiming?

Public engagement counts cannot establish campaign ROI, spend, revenue attribution, impressions or unique reach. The skill says to report these as unavailable unless you supply authorised first-party data.

Follower growth needs repeated, timestamped observations, since a single profile snapshot cannot show growth. The skill also asks the agent to state the engagement-rate denominator, observation window and collection time, and to deduplicate by stable content ID.

  • Compare like formats with like formats, and like platform metrics with each other
  • Keep missing values distinct from zero
  • Use a single snapshot for comparison only, not for growth claims

What people use it for

  • Comparing engagement on recent Instagram reels to see which formats perform best
  • Reviewing which YouTube Shorts or TikTok videos drew the most views over a set period
  • Exporting a bounded, timestamped set of Facebook post metrics for a monthly content review
  • Auditing a brand's public hashtag performance on Instagram before planning posts
  • Pulling Facebook ad reach-related public data for a creative comparison, with the limits above applied

Frequently asked questions

Is the apify-content-analytics skill free?

The skill's licence is MIT. Running it, however, uses Apify, and the skill notes that each run can incur charges, so you need an Apify account with credits. Check the pricing of the chosen Actor before launching.

What do I need to install the Apify content analytics skill?

Download the ZIP from this page and unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder, or upload it in the Claude apps under Settings → Capabilities → Skills. You also need Node.js 20.6 or newer and an APIFY_TOKEN in a local .env file.

Can this skill measure campaign ROI or ad spend?

No. The skill states that public engagement counts alone cannot establish campaign ROI, spend, revenue attribution, impressions or unique reach. Those figures need authorised first-party data, which you would supply yourself.

Which platforms does it support?

The skill's Actor list covers Instagram, Facebook, YouTube (including Shorts) and TikTok. Each platform uses a different scraper, and the agent should confirm the current Actor and its input schema before running.

Why is my run failing with APIFY_TOKEN not found?

The script reads the token from a local .env file. Create .env in the folder you run from, add APIFY_TOKEN=your_token, and make sure you run the command with --env-file=.env. Do not paste the token into chat.

Can it collect private group data or personal contact details?

No. The skill forbids collecting private group data and unnecessary personal contact information, and it advises redacting personal fields before sharing. Public data only, within platform terms.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

Use this skill if you are a marketer, creator or analyst who wants repeatable, bounded exports of public engagement data across major platforms and can accept paid Apify runs. Pick something else if you need true campaign ROI, ad spend or revenue figures, or if you need private or contact-level data, since the skill rules those out.

Download the skill