Collect Amazon Product data and return results
Publisher
Skills by dataify-server
80 published skills, 657 installs in total.
Collect Amazon Global Product data and return results
Collect Reddit Posts data and return results
Collect eBay Products data and return results
Collect Facebook Comment By URL data and return results
Collect Facebook Profile By URL data and return results
Collect Instagram Comment By Post URL data and return results
Collect Instagram Profiles data and return results
Collect Instagram Reels data and return results
Collect Facebook Events data and return results
Collect Facebook Post By URL data and return results
Collect Google Map Details data and return results
Collect Reddit Comment By URL data and return results
Collect Walmart Products data and return results
Collect YouTube Profiles data and return results
Search Google Shopping for products
Search Google Local for nearby places
Collect Google Maps Reviews data and return results
Collect YouTube Audio By URL data and return results
Collect YouTube Comment By ID data and return results
Collect YouTube Product By ID data and return results
Collect YouTube Transcript By ID data and return results
Collect YouTube Video By URL data and return results
Collect YouTube Video Post data and return results
Collect Indeed Companies Info data and return results
Unlock blocked or JS-heavy pages through Dataify
Collect Amazon Product List data and return results
Search Google Finance for financial data
Search Google Images
Search by image using Google Lens
Search Google Maps for locations
Search Google Patents
Search Google Scholar for academic papers
Search Google Trends for trend data
Search Google Flights for flight prices
Search Google Jobs for job listings
Search Google Play Store for apps
Search Google Videos
Collect Airbnb Builder data and return results
Collect Amazon Comment data and return results
Collect Amazon Seller data and return results
Collect Booking Hotel Info data and return results
Collect Crunchbase Builder data and return results
Search Google Hotels for hotel prices
Search Bing Images
Collect Github Builder data and return results
Collect Glassdoor Builder data and return results
Collect Linkedin Builder data and return results
About Ottermind skills
Every published skill from dataify-server. Following a publisher whose work already fits how you operate is often faster than evaluating individual skills one at a time.
Skills cover the full range of what an Ottermind agent does: building editable PowerPoint decks, generating product and brand images, writing and running code on a machine you connect, producing video from a script or a still, and cleaning and analysing spreadsheet data. A skill is what turns a general model into something that handles your specific version of one of those jobs the same way every time.
A skill is a plain Markdown file with a short frontmatter header, which means you can read exactly what it will do before running it, edit it to match your own conventions, and share it with colleagues. Once a skill produces the result you wanted, it can be attached to an agent or scheduled as an automation so the same work happens again without anyone rewriting the prompt.
- What is an Ottermind skill?
- A reusable instruction package written in Markdown. It gives an agent the context, steps, and constraints for one kind of task, so results stay consistent instead of depending on how well you phrased the prompt that day.
- How do I install and use a skill?
- Open any skill's detail page, read the full document, and install it into your workspace. From then on the agent can use it automatically when the task matches, or you can invoke it by name.
- Can I create my own skill?
- Yes. A skill is a Markdown file with frontmatter naming it and describing when it should be used. If you have a workflow you keep re-explaining to an agent, writing it down once as a skill is usually the shortest path to making it repeatable.
- Do skills cost extra?
- Browsing and installing skills from this library does not cost anything on its own. Running one consumes the same credits as any other agent task, depending on the model you choose and how much work the task involves.