Hyperterse

Give your agents structured access to your databases in minutes, not months, and make them work for you. Here’s a quick demo of what Hyperterse can do for you ⬇️

Spent a lot of time tightening the core loop in this one. The goal is simple: You shouldn’t be writing MCP servers. You should be declaring intent. v2.3.0 gets us meaningfully closer. - SQLite connector now natively supported - Ability to create prompts, resources and templates

⚡️ v2.3.0 just dropped. The goal is simple: You shouldn’t build MCP servers anymore. You should just describe them. Everything else? Handled - Full adherence to MCP spec - SQLite connector Upgrade: #hyperterse-upgrade" target="_blank" rel="nofollow noopener">docs.hyperterse.com/reference/cli#… Full changelog: github.com/hyperterse/hyp…

I’m so excited for what’s next with Hyperterse. With a simple config, you can basically build your own agents and extend them Have a demo that’s coming along nicely!

v2.2.1 is out! Agent Skills folder generated correctly when initialising a new Hyperterse project. Upgrade: #hyperterse-upgrade" target="_blank" rel="nofollow noopener">docs.hyperterse.com/reference/cli#… Full changelog: github.com/hyperterse/hyp…

We have released v2.2.0. This contains an important security fix wherein the `search` tool exposed raw unsubstituted database statements. The relevance score when searching has also been improved. Upgrade: #hyperterse-upgrade" target="_blank" rel="nofollow noopener">docs.hyperterse.com/reference/cli#… Full changelog: github.com/hyperterse/hyp…

I have built a few hundred agents over the last year, and it’s always been the same start - choosing the right SDK for the job. I’m building @hyperterse to solve this issue for MCPs and agents alike. With security in mind.

This is something that I am most excited about. Especially security. Right now, head down and building the ability to create agents declaratively!

The direction we are choosing to move towards is all about taking care of as much stuff as possible. Doing the heavy lifting for you. You just focus on your business logic.

Security One problem with building enterprise MCPs is authentication + authorization. Hyperterse was built on security as on of the pillars and the goal was always to eventually address it. Starting now, there will be regular security-focused patches and increments.

The next few versions will focus on some aspects which unimportant for the framework to grow: - Bring your own sandbox - Agents - Security

We've just released v2.1.0. With this release, when you initialize a new Hyperterse project it automatically adds an agent skill to help your agent build your MCP server! Upgrade: #hyperterse-upgrade" target="_blank" rel="nofollow noopener">docs.hyperterse.com/reference/cli#… Full release notes: github.com/hyperterse/hyp…

I'm building @hyperterse, the framework for MCPs. I'm looking for a founding GTM partner who loves selling dev infrastructure and talking to engineers. If you've taken a devtool or infra product from 0 → first customers, DM me.

AI space is evolving fast, with a framework you don't need to worry about the plumbing. Focus on moving just as fast.

Hyperterse v2.0.0 just dropped. Stop duct-taping MCP servers. Hyperterse is the framework for building secure MCP tools for your agents. Only two tools, search & execute → agents get real access. No glue code. No hallucinated data. hyperterse.com/?ref=x

Coming soon to a browser near you!

We've just released v1.4.0 with an intent to safeguard your databases against a query-barrage! With this release, you can now setup local caching at the runtime layer. Upgrade: #hyperterse-upgrade" target="_blank" rel="nofollow noopener">docs.hyperterse.com/reference/cli/… Full release notes: github.com/hyperterse/hyp…

Agents handing off tasks across clouds, databases, Slack is the future we all want. But handoffs without structure are just distributed guesswork. Agentic systems don’t fail because they’re autonomous. They fail because nobody defined the contract properly.

The real question is - is schema awareness the only bottleneck to minimize hallucinations in text-to-SQL? We’ve built production agents, and schema awareness solves one problem, but still doesn’t paint a complete picture. Here’s where we leverage MCP to solve this ambiguity.