This product was not featured by Product Hunt yet.
It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).

Product upvotes vs the next 3

Waiting for data. Loading

Product comments vs the next 3

Waiting for data. Loading

Product upvote speed vs the next 3

Waiting for data. Loading

Product upvotes and comments

Waiting for data. Loading

Product vs the next 3

Loading

Pyas

Calendar API v2, now with MCP and agent tools

One API for Google Calendar, Outlook and Zoom. Pyas v2 gives Claude, Cursor or your own agent your users' calendars. Add the hosted MCP server by its URL and sign in, or use the same tools with the Vercel AI SDK, LangChain, OpenAI or Anthropic. Agents find meeting times, create events with Meet, Teams or Zoom links and answer invitations. Google, Microsoft and Zoom have already verified Pyas's sign-in, so you skip the OAuth review. Pyas runs no AI models.

Top comment

Hi Product Hunt,

Pyas launched here as a calendar API for Google, Outlook and Zoom, and it has been running in production since 2023. Everything those years taught us about keeping real calendars in sync went into v2. It's out today, and the biggest change is that your agent can use it directly.

Add https://mcp.pyas.io/sandbox to Claude, Cursor or VS Code and sign in. No key to paste. Then ask "find 30 minutes with Sam next week and send him an invite with a with a Meet link". The agent calls find_meeting_times, then create_event, on sample calendars in your app's sandbox. When you want real calendars, add https://mcp.pyas.io as well.

There are three ways to use it:

  • Assistants. Claude, Cursor, VS Code or any MCP host can read and change events, find free time, answer invitations and add meeting links.

  • Coding agents building on Pyas. The API reference, error codes and your request logs are tools, so "why did my last request fail?" gets an answer from the actual trace.

  • The agent inside your product. @pyas/ai-tools gives you the same tools for the Vercel AI SDK, LangChain, OpenAI or Anthropic, so your agent can schedule for your users.

Calendars hold real people's days, so the server is careful:

  • Every tool says whether it reads, changes or deletes, so the host can ask you first. The npm package's --read-only flag offers only tools that change nothing.

  • Event titles and descriptions are written by other people. The server marks them as data, so an agent doesn't take orders from an invitation.

  • Test keys reach only the sandbox. On live calendars, you confirm each change.

  • Plans, payment and live keys never go through the agent. It sends you a dashboard link.

Pyas runs no AI models. Your calendar data isn't sent to a model provider by us, and there's no per-call AI cost on your bill.

Underneath is the new v2 API: one event shape for Google and Outlook, signed webhooks retried for 41 hours, a TypeScript SDK, a CLI and a new dashboard. Your users still connect through Pyas Auth, which Google, Microsoft and Zoom have already verified, so you skip Google's OAuth review and Zoom's app review. When you want your own name on the consent screen, switch any provider to Native Auth and use your own OAuth app. If you're on v1, nothing changes until you move to v2, and your keys and connected accounts carry over.


You can find the docs here: https://developer.pyas.io/ai-tools

If you're building with agents, I'd like to hear what you'd want one to do with a calendar.

About Pyas on Product Hunt

“Calendar API v2, now with MCP and agent tools”

Pyas was submitted on Product Hunt and earned 2 upvotes and 1 comments, placing #156 on the daily leaderboard. One API for Google Calendar, Outlook and Zoom. Pyas v2 gives Claude, Cursor or your own agent your users' calendars. Add the hosted MCP server by its URL and sign in, or use the same tools with the Vercel AI SDK, LangChain, OpenAI or Anthropic. Agents find meeting times, create events with Meet, Teams or Zoom links and answer invitations. Google, Microsoft and Zoom have already verified Pyas's sign-in, so you skip the OAuth review. Pyas runs no AI models.

On the analytics side, Pyas competes within Calendar, Developer Tools and Artificial Intelligence — topics that collectively have 1M followers on Product Hunt. The dashboard above tracks how Pyas performed against the three products that launched closest to it on the same day.

Who hunted Pyas?

Pyas was hunted by Freinet Brutus. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.

For a complete overview of Pyas including community comment highlights and product details, visit the product overview.