How to use the Brand Reputation Monitoring agent skill

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

An agent skill is a folder containing a SKILL.md file of instructions, plus optional scripts and reference files, that an AI agent loads when a task matches the skill's description. This skill's name is apify-brand-reputation-monitoring, and its description is to scrape reviews, ratings and brand mentions from multiple platforms using Apify Actors.

Once installed, the agent can select a suitable Actor for the platform you name, check its input schema, run it through the bundled script reference/scripts/run_actor.js, and save the results to a CSV or JSON file or show a few records in chat. It then reports the number of records, the file location, the key fields and suggested next steps such as sentiment analysis or filtering.

What you need before installing

The skill has real requirements. Check these first, because paid cloud runs cannot work without them.

• Node.js 20.6 or newer (the script relies on native --env-file support).
• An Apify account and API token, stored in a local .env file as APIFY_TOKEN.
• Apify paid Actors may incur usage charges. The skill says to verify prerequisites before any paid run.
• Optional: an Apify connector or Apify Console access, so the agent can inspect Actor schemas.

Install the skill: download, unzip and place it

Download the ZIP for this skill from the page you are reading. Unzip it, then put the skill folder in one of two places depending on how you use your agent. A personal install goes in ~/.claude/skills/. A project-only install goes in the .claude/skills/ folder inside your project.

If you use the Claude apps instead of files on disk, you can upload the skill under Settings → Capabilities → Skills.

• Personal: ~/.claude/skills/apify-brand-reputation-monitoring/
• Project: .claude/skills/apify-brand-reputation-monitoring/ inside your project folder
• Claude apps: Settings → Capabilities → Skills, then upload the skill

Set up your Apify token safely

Create a file named .env in the directory where you run the agent and add your token as APIFY_TOKEN=your_token. The skill's error handling asks for exactly this line if the token is missing.

Keep the .env file private and out of version control. The skill says to configure the token locally and not paste it into chat or into command arguments.

How to trigger the skill with natural requests

You do not need to type a command. Describe the job in plain language and name the platform and the brand or place. Examples of requests that match the skill's scope include asking for recent Google Maps reviews of a named business, Booking.com reviews for a named hotel, TripAdvisor reviews for a named restaurant, or Instagram or TikTok comments and hashtag or tag mentions for a brand.

The agent will then ask you two things before it runs: the output format (quick answer in chat, CSV, or JSON) and the number of results you want. Be specific here. A small limit is safer for a first test.

• Pull the latest 50 Google Maps reviews for [business name] and save them as CSV.
• Show me a few recent TripAdvisor reviews for [restaurant] in chat only.
• Export Instagram comments on our last three brand posts to JSON for a human sentiment review.

Example run and what the output looks like

For a quick look with no file saved, the agent runs the script with an Actor ID and a JSON input. Quick answers fetch at most five records. For a saved export, the agent adds an output file and a format. The skill requires output file names to be new paths and suggests a date prefix in the form YYYY-MM-DD.

The skill's Actor table maps needs to Actor IDs, for example compass/crawler-google-places for Google Maps business reviews and clockworks/tiktok-comments-scraper for TikTok comments. These are discovery candidates. The agent must still confirm the Actor's current schema before running it.

Safety notes: scripts, network and private data

This skill runs scripts and uses the network. The helper script sends requests to Apify and sends a skill-identifying User-Agent header. Before a launch, the skill says the agent should confirm the Actor, the authorised public-data scope, the input schema, the result limits and the pricing with you.

The skill sets clear boundaries. It does not collect private group data or unnecessary personal contact details, and it says to respect source terms and access controls. It does not post replies, contact reviewers or remove content. If a run times out, the local timeout does not stop the remote run, so check the printed run ID in Apify Console before retrying to avoid duplicate charges.

• --max-items limits only the downloaded output (default 1000). It does not cap Actor work or billing.
• Minimise or redact personal fields before sharing any exported data.
• If a run shows FAILED, use the Apify Console link in the error output.

Tips for cleaner results

The skill asks the agent to deduplicate reviews by platform and stable ID, and to record the rating scale and retrieval time. It also tells the agent to keep the source's own rating separate from any inferred sentiment.

Counts from different platforms are not directly comparable unless you define coverage and scale. Keep negative, neutral and uncertain items in the data. Before you label a trend, have a person check a small sample for language and sarcasm.

• Start with a small input limit to confirm the Actor and cost.
• Treat each export as a snapshot. The skill does not monitor continuously.
• Scheduling or alerts need a separately configured service that you set up yourself.