MC Project Lab
I develop software for enterprises and professionals who use artificial intelligence and need to control what it does, what data it processes, and what evidence supports its outputs.
The laboratory tools restrict AI agent actions, detect sensitive data, and verify answers before they are used in operational workflows. I engineer and stress-test them through technical audits and authorized attack simulations.
Some solutions operate locally, while others integrate with cloud services. For each product, I state what is available, what has been verified, and what remains under development.
Technologies and Integrations
The laboratory works with Google Cloud Run, Cyrus CLI, and Android Keystore. Available integrations and their operational boundaries are documented in the product specifications.
AI Agent Security
I develop tools to stress-test autonomous agents and evaluate how they react to malicious instructions, out-of-scope requests, and data exfiltration attempts.
Geronimo, Medusa & Ghost Assailant
Tools to simulate authorized attacks in controlled environments. Used to identify unexpected behaviors and verify security controls.
Ghost Assailant 路 Geronimo 路 Medusa AI
Explore Red Team →AIPF, Vigil & Maskyan
Tools to analyze requests, enforce boundaries, and protect information processed by AI systems. Each product has a defined scope of use and verification.
AIPF 路 Vigil 路 Maskyan
Explore Defense Lab →AgentScoope & Armor Suite
Tools to monitor agent activities, inspect logs and files, and verify the integrity of collected evidence.
AgentScoope 路 Armor Suite
View Lab Results →Breakpoint
In Episode 09 we discuss when an agent must stop. Even with authorized tools, agents require strict limits on activities, time, and compute budget, alongside a human operator empowered to intervene.
Who is in the Lab
I am Maurizio Cirio, founder of MC Project Lab. I oversee project selection, engineering, and verification, supported by AI tools.
Eva Kane, Max Voss, NYX, and Human are virtual personas used to present news, analyses, and projects. They are not human employees.
Enter the Studio →
Explore Studio Virtual Profiles
Laboratory Activities
From engineering the control tower for AI agents to authorized security stress tests and public observatories.
TCT
TCT brings AI agent coordination, activity controls, and human-in-the-loop approvals into a single unified environment.
Discover TCT →In the Lab
Inside the laboratory I conduct authorized tests, collect results, and document what worked, what did not, and under what constraints.
Open the Lab →Sources & Data
Three public websites dedicated to energy, robotics, and space. They aggregate sources, market data, and updates to make them easier to analyze.
Explore All Products →Investors and Partners
Seeking partners to bring laboratory products to market, establish commercial priorities, and build the required commercial and technical capabilities.
Integrate a laboratory technology into your products or services.
Support the development and commercial scaling of a project under agreed terms and structure.
Test a solution within your enterprise or institution, starting from a concrete business need and measurable outcomes.
Distinguish what AI measured, read, or inferred from what it recalled without fresh verification. Tie outputs to available evidence so decision-makers can evaluate them.
Observatories
Three public websites aggregating sources, data, and updates to make them easier to consult.
-
/// ENERGY Energy ObservatoryVisit Website →
Energy prices, fuel benchmarks, utility bills, and market updates.
-
/// ROBOTICS Robotics ObservatoryVisit Website →
Robotics, humanoids, and physical AI applications.
-
/// SPACE Space ObservatoryVisit Website →
Satellites, rocket launches, commercial space ventures, and space regulations.
Let's Talk
To explore a product, propose a partnership, or evaluate a pilot deployment in your organization, write to me.
Maurizio Cirio
info@mcprojectlab.com
mcprojectlab@proton.me