Privacy Policy

This policy explains what skillsandagents.co (the "Site") collects when you visit, how it is used, and who processes it. It covers the Site itself. It does not cover the skills, agents, or code you download and run on your own systems — the use of that software is governed by our Terms of Service and the license stated for each item (some are released under the MIT License, others under their own licenses such as Apache-2.0). Those terms address the software, not your privacy as a visitor to this Site.

What we collect

The Site is primarily a static catalog. You do not need an account to browse it. We do collect basic, privacy-respecting usage analytics: pages viewed, sessions, file downloads, and interactions with elements on the page, along with the standard technical data a browser sends (such as your IP address, device, and browser type). This is handled by the processors listed below.

By default, this analytics runs anonymously: we don't store anything on your device, and we don't link your visits together. This is true before you make a choice in the cookie banner and if you click Decline. If your browser has Do Not Track turned on, we skip this anonymous tracking too, and collect nothing. If you click Accept, we switch to fuller tracking that does store an identifier in your browser so we can recognize repeat visits (see "Cookies and local storage" below).

Information you send us

The Site has a Contribute form where you can submit a skill or agent for the catalog. If you use it, we collect your name, email address, and the GitHub repository URL you provide. We store this information in Cloudflare KV and use it only to review your submission and contact you about it. We do not sell it. You can ask us to delete it at any time (see "Data requests" below).

The Site also has a Contact form. If you use it, we collect your name, email address, and the message you write. We store this information in Cloudflare KV and use it only to read and respond to your message. We do not sell it. You can ask us to delete it at any time (see "Data requests" below).

When you download a skill or agent, the Site shows a panel with a subscribe form. If you use it, we collect your email address, the page and skill you subscribed from (the source path and slug), and a timestamp. We store this information in Cloudflare KV and use it only to email you about new skills and updates. We do not sell it. You can ask us to delete it at any time (see "Data requests" below).

Several places on the Site also invite you to email us, for example the Adopt AI page and the contact us links. If you choose to write to us, we receive your email address and whatever you include in your message, and we use it only to respond to you and to follow up about the work or request you raised. We keep that correspondence for as long as needed for that purpose and our records, and we do not sell it. You can ask us to delete it at any time (see "Data requests" below).

The Site also has a programmatic agent-discovery search endpoint that AI clients use to search the catalog. When a search is submitted through it, we store the search text along with the number of results it returned, any filters applied, the country, and a timestamp. We store this in Cloudflare KV, it expires automatically after 90 days, and we use it to see which searches don't return a good match so we know what skills to build next. We do not sell it. (The search box on the catalog page itself filters in your browser and does not send your search text to us.)

How we use it

We use analytics data, and the searches submitted through the agent-discovery endpoint, to understand which skills and agents people find useful, to see where the catalog is missing something people are looking for, to fix problems, and to improve the Site and the skills we offer. We may review how the platform is used, including that search activity, to improve the product experience and the skills we provide. We use anything you send us to reply and to handle your request. We do not sell any of it.

Processors

We rely on a small number of third-party services to run the Site:

• Amplitude — product analytics. By default it runs anonymously: it records page views and sessions, and processes the technical data described above, but doesn't store anything on your device and doesn't link your visits together. This anonymous mode runs before you decide on the cookie banner and if you click Decline, unless your browser has Do Not Track turned on, in which case it doesn't run at all. If you click Accept, it also tracks file downloads and form-style interactions, and stores persistent identifiers (such as device and session IDs) in your browser's cookies and local storage so it can recognize repeat visits. The analytics script loads from a third-party CDN (jsDelivr).
• jsDelivr — content delivery network. When analytics is enabled, your browser fetches the Amplitude SDK from jsDelivr's CDN before Amplitude initializes; jsDelivr receives the request data involved in serving that file, such as your IP address and user agent.
• Google Fonts — the Site loads its typeface from Google's font servers (fonts.googleapis.com and fonts.gstatic.com). Google receives your IP address as part of serving the font files.
• Cloudflare Pages — hosting. Cloudflare serves the Site and processes request data, including your IP address, as part of delivering pages to you.
• Google Ads — conversion tracking. We use Google Ads to measure conversions from marketing campaigns. Google stores conversion-tracking cookies in your browser (such as _gcl_au) to attribute user actions back to ads you may have seen.

Each of these providers processes data under its own privacy policy.

Cookies and local storage

The first time you visit, you'll see a banner asking whether you're okay with analytics and ad tracking. Your choice is saved in your browser's local storage. Google Ads never loads until you accept, and no conversion-tracking cookies are set until then. Amplitude works differently: it runs in an anonymous mode by default, with nothing stored on your device, before you decide and if you decline — unless your browser has Do Not Track turned on, in which case it doesn't run at all. If you accept, Amplitude switches to storing persistent identifiers — for example device and session IDs — in your browser's cookies and local storage, so it can recognize repeat visits and stitch together a session, and Google Ads starts storing conversion-tracking cookies in your browser (such as _gcl_au) to measure campaign performance.

This choice only covers analytics and ad tracking. It does not cover every third-party request the Site makes — for example, the Site's fonts load from Google's font servers regardless of your choice, and Google receives your IP address as part of serving those files (see "Processors" above).

You can change your mind at any time by clearingcookieConsent from your browser's local storage, which brings the banner back. You can also clear cookies and local storage entirely through your browser's settings, or block them with your browser's cookie and storage controls; doing so does not affect your ability to use the Site.

Data requests

If you want to ask what data relates to you, or request its deletion, email contact us and we will do our best to help.

Changes

We may update this policy from time to time. When we do, we will revise the "Last updated" date above.

Contact

Questions about privacy? Email contact us.