Mintive

Infrastructure for the agentic enterprise. MCP & LLM gateways. Deploy agents with identity and permissions. (fka MintMCP)

[Mintive @mintive]

@AmirMushich @yulikay Thanks @AmirMushich !

[Mintive @mintive]

@ModernTreasury put a bank-ops agent live on Mintive in under three weeks. A four-hour task now takes minutes.

@UMich runs Mintive for 150,000 students, faculty and staff.

@coursera, @Braze, @ro, @StabilityAI and @apax_partners run on it too.

https://t.co/1Jbl9boZ8p

[Mintive @mintive]

AI agents are becoming part of how enterprises operate. They need identities, permissions and security built for the scale ahead.

Today, Mintive announced $11M in funding, 50+ paying customers and 2.5M+ daily agent calls.

We're building the infrastructure for the agentic enterprise.

[Jiquan Ngiam @JiquanNgiam]

Every AI agent at work runs with the access of whoever set it up.

That’s why security keeps saying no.

Today we’re launching @mintive: every agent gets its own identity and permissions, the same way every employee does.

We’ve raised $11M, with @cowboyvc leading our latest round.

[Mintive @mintive]

Every AI rollout has two owners. Enablement wants agents for everyone. Security has to sign off.

With Mintive, enablement connects Claude, ChatGPT, Copilot, Cursor and in-house agents to company data, and security runs one policy across all of them.

[Mintive @mintive]

@trycaliber @JiquanNgiam @CowboyVC That's why have agent identities and their own permissions that live beyond the any employee's departure. The agent will have a good definition of their permission and purpose so it can keep functioning!

[Mintive @mintive]

@markkmii @JiquanNgiam @CowboyVC Great questions!

IT set up the base connectors (gmail, CRM, github, etc) for the entire company. The team that owns the work will compose what connection they need for the agent.

More details: https://t.co/8S9c5KhDQc

[Mintive @mintive]

@abarrallen we're grateful for your support @abarrallen!

[Mintive @mintive]

@AndrewYang33 Thanks @AndrewYang33 ! We have more than a few customers come back to us months after our initial calls come to the same conclusion. You can't vibe-code security :)

[Mintive @mintive]

@ericbahn @JiquanNgiam @HustleFundVC thank you for your partnership!

[Mintive @mintive]

@ethanxpy @MAXdeg0 Yes! For our Coworker agent product, we use the caller's identity as the true identity passed into downstream calls to fix this problem

[Mintive @mintive]

@Harish_521 @HelloVyom this is one of the problems we are solving!

[Mintive @mintive]

@harleyfoote_ @slash1sol We also have auto Prompt Injection blocking from MCP tool returns with our guardrail product Mint Guard (https://t.co/UXV5Be3v2p) which will auto scan the request and response of MCP tool calls

[Mintive @mintive]

@harleyfoote_ @slash1sol Exactly.
Our architecture design give you the ability to freeze and version the tool and block auto-tool description update, or even override individual tool description to avoid these type of supply-chain attack https://t.co/83YNSTiUvU

[Mintive @mintive]

@raulnogales @JiquanNgiam @CowboyVC We have an agent identity product https://t.co/RKuZSX9VXH that can both audit and block calls that's not supposed to happend.

With our MCP gateway, we provide fine-grain control (turn off individual tools, or rewrite tool description, record caller identity and more)

[Mintive @mintive]

Great question.

You can configure agents to act on behalf of someone who called them. When the agent takes an action, we check that the caller still has access, e.g. if someone asks an agent to update a spreadsheet overnight but that person loses permission to update the spreadsheet before the agent runs, the agent's attempt to update the spreadsheet will be denied.

[Mintive @mintive]

@raulnogales @JiquanNgiam @CowboyVC exactly the problem we're tackling!

[Mintive @mintive]

@dunkhippo33 @JiquanNgiam we're so grateful for your support!

[Mintive @mintive]

@roadtoCISO @JiquanNgiam @CowboyVC honored to work with you and your team @roadtoCISO!