Meko is Open for Self Sign-Up: Here’s How to Get Started!

AI agents are great at reasoning, but they’re forgetful. You can coach one through the shape of a tricky migration in Cursor on Monday, and by the time you open Claude Code on Tuesday, you’re explaining it all over again.

Every tool keeps its own memory, and none of them talk to each other.

Meko was built to fix that problem.

We are pleased to share that Meko is now open for self sign-up! This blog explains how to create a free account today to start giving your agents persistent, shareable memory.

What is Meko?

Meko is an agent-native data layer that lives independently of any single AI vendor.

Instead of your context being trapped inside one chat client or coding harness, it lives in Meko and follows you wherever you go. Because it connects over MCP, you can plug it into whatever you already use, including Claude, ChatGPT, Cursor, and other MCP-compatible tools.

The core idea is simple: your agents get a memory that persists across sessions, tools, and time, plus a searchable knowledge base and a full record of how they reached their answers.

Meko is private by default, and you choose what to share and when.

What Does Meko Do?

Everything in Meko lives in a datapack. This is a container for a project or workstream that holds the four things agents actually need:

• Conversations: The full interaction history between you and your agent.
• Memories: The takeaways your agent extracts from those conversations, scoped privately to each agent by default.
• Shared knowledge: Documents you upload (PDFs, markdown, text files) that Meko chunks, indexes, and makes searchable for any agent with access.
• Decision traces: The reasoning behind an answer, down to the tools the agent used on each turn, so you can see why it responded the way it did.

A few things this unlocks in practice:

Your context survives switching tools

Build up a week of project context in one client, switch to another, and ask, “what do you know about my current project?” As long as both point to the same datapack, the answer comes with you.

You share what your agent learned (without sharing everything)

Rather than an all-or-nothing memory store, Meko lets you promote specific memories into a shared layer while keeping your raw conversations and per-agent noise private. Your team gets the distilled insight without wading through every draft and dead end.

You onboard people, not just their agents

When you share a datapack, a teammate’s agent has the project context from day one, including a window into how the work was done, not just the final result.

How to Get Started

Self sign-up for Meko is now open!

Anyone can create a free account: just click on the ‘Sign Up’ button on the Meko homepage.

The free tier includes:

• 10 datapacks
• 1,000 conversations per month
• 10,000 retrievals per month
• 10M workbench tokens
• 100MB knowledge storage (5MB max file size)

Once you’re in, you can:

• Connect any MCP client: Claude, ChatGPT, Cursor, and more.
• Save and recall context in one place, no matter which AI tool you’re using.
• Upload docs into your own knowledge base for your agents to search.
• See why your agent answered the way it did: the reasoning and tools used on each query.
• Get community support in the Meko channel on the YugabyteDB Discord.

You can request an upgrade to Pro inside the Meko portal if you hit limits and want to go further.

Try Meko Today

Sign Up for Meko, and connect your first agent.

Share any comments/issues/tips in the #early-access channel on the YugabyteDB Discord server.

Find out more about Meko by reading these recent blogs:

• Discover Meko: The Data Infrastructure for Agents That Work and Learn Together
• Get Started with Meko: Agent Memory with Built-in Discernment

We’re building Meko in the open, and the more you use it (and share feedback), the faster we can make it better. Thank you for joining us on this journey!