Virtual MCPs was a concept our team invented, more than a year back! The key idea is to have MCPs that aggregate across connectors and be use case driven. This is more relevant now than ever.
Read more here: pulsemcp.com/posts/virtual-m…
Our latest framework on how to think about securing agents in the enterprise.
mintive.com/blog/treat-ai-ag…
Phishing is getting smarter, more personalized, and harder to spot. This one knew exactly how to get our CEO's attention.
Just got hit with a pretty sophisticated phishing attempt.
Someone posing as a Bloomberg contact reached out about a conversation I'd genuinely be interested in. Convincing profile, blue checkmark on X, and a well-targeted pitch.
They definitely did their homework.
The red flags:
--Their scheduling link redirected to an X OAuth authorization page requesting full access to my X account.
--I Googled the person and couldn't find anything about them.
--I suggested using my own Calendly link instead. No response.
Didn't fall for it. But phishing is getting very targeted. A blue checkmark means nothing. Always check what access you're granting.
And @business, if you actually want to chat about Mintive's recent raise and how we're securing enterprise AI agents, I'm all ears 😄
3
282
Mintive retweeted
Replying to @spenserskates
This keeps me up at night. The biggest problem now isn’t adopting AI anymore. It is securing AI - both from it and how we use it.
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.
@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.
mintive.com/case-studies
46
Read more here: techcrunch.com/2026/10/29/ag…
We’re excited to partner with great investors @CowboyVC, @OpCo_VC, Burst Capital, @coatuemgmt, @HustleFundVC and @mavenvc.
We’re grateful to have angels supporting us: @fkerrest (Okta), @grinich (WorkOS), @scottbelsky, @JeffDean and @karpathy.
6
93
You get an agent!
... and you get an agent!!
... and YOU get an agent!!!!
Had a blast at the first AARM event hosted by Herman Errico at @TrustVanta offices. Did some cool demos on how we're approaching permissions for multiplayer agents, showing how we need to use the context around the channels, requester, and tools to determine what agents can actually do!
11
19
492
Mintive retweeted
How do you build an AI agent that can investigate customer issues across docs, code, production data, and incident history?
Ankit, our Engineering Manager, shares how we built a Bank Operations Agent to orchestrate our internal tools and save our Technical Account Managers hours on support cases.
He covers:
- How the agent decomposes requests and queries multiple systems in parallel
- Why docs, code, production data, and runtime signals each have distinct epistemic roles
- How @mintive enables a single agent to orchestrate workflows across our internal toolchain
- Why grounding and human review matter more than maximizing autonomy
Read more: go.moderntreasury.com/4pDLkq…
Mintive retweeted
i didn't realize how quickly the agentic supply chain has increased the attack surface for prompt injection until speaking with @JiquanNgiam at @mintive
MCPs make building so much easier, but over 30% are vulnerable, giving attackers an easy way into internal systems. skills, memory, and local cli can all be easy paths in.
while many teams are thinking about "how do i get my team to adopt ai" the best teams have already adopted and are being intentional about using a gateway as a control to manage permissions, MCP access and hooks to reduce vulnerabilities.
Mintive retweeted
MCP Apps are a better replacement for elicitation
We build an enterprise MCP gateway, and on first use of a service, users have to authenticate with the underlying MCP service. MCP apps can be a rich, in context experience that supplants elicitation.
Mintive retweeted
When we ask an agent to "fix this bug," we're approving hundreds of actions we'll never see. These agents have our production credentials; and increasingly, they run in the background while we work on something else.
Engineers are spinning up multiple Claude Code sessions in parallel, and ClawdBot hit >100k GitHub stars in weeks by letting agents handle email, workflows, even car purchases. The shift to agents without constant supervision is here.
Today we're launching @mintive : governance for AI agents.
Mintive retweeted
Cursor and MintMCP are partnering to help enterprises govern their coding agents through Cursor's new hooks program.
Engineering orgs are rapidly increasing their use of Model Context Protocol in coding workflows. But most orgs have no visibility into what's actually installed, what data flows through these connections, or how to govern them.
Cursor's hooks enable you to observe and control the agent loop by intervening at specific points. With MintMCP, you can enable hooks that fire before and after MCP tools are used.
After enabling these hooks, you get:
🔹 A full inventory of all MCP servers installed across your organization
🔹 Allow/deny policies on servers and tools
🔹 Response scanning for sensitive data before it reaches the AI model
🔹 Complete audit trails for compliance
Start by collecting data to understand actual usage patterns, then layer governance on top.
Check out Cursor's announcement and our blog post on MCP governance using hooks - links in the comments below.
@cursor_ai @MCP_Community @mintive














