qashqade, the leader in allocation management, today announced a new set of MCPs (Model Context Protocol connections) that let clients connect their own AI tools directly to qashqade. Through them, the AI a team already uses can work with qashqade in plain language, bringing the platform into the AI workflows clients are already building.
Rather than opening the interface to find a figure, a client’s AI can now ask qashqade a question, trigger a calculation, and read back validated results, then carry the answer straight into whatever tool the team is already working in. An accountant can ask how much carry to book for a vehicle; management can ask for the total distribution across a portfolio; investor relations can pull a partner’s realised gain to date; all without logging into qashqade to extract it by hand. In effect, qashqade becomes one more tool a client’s AI can call on, on demand, inside their own stack.
What does not change is where the numbers come from. The AI never performs the allocation math. It asks qashqade’s deterministic engine to run and reads back what the engine produced, so every figure is computed the same way on every run, traceable to its source and defensible to LPs and auditors. The calculation core stays exactly as it is today; an MCP is simply a second door into it: the interface is the door a person uses, an MCP is the door an AI uses. An AI also signs in as a known qashqade admin user and inherits that user’s permissions, so it never sees more than the person behind it.
The MCPs work with any MCP-compatible AI, ship with ready-made skills so a client’s AI knows how to use them, and are available today, for existing and new clients alike. Give us a call so that we can switch it on for you.
Commenting on the launch, Oliver Freigang, CEO of qashqade, said:
“Almost every firm we talk to is busy connecting AI to their systems, and there is one question they all come back to: when an AI hands you a number, can you trust it to be right every single time? We have been stubborn for years about keeping our allocation engine deterministic (same inputs, same outputs, every run) and that discipline is exactly what lets us do this with confidence. Clients can now bring qashqade into their AI workflow and pull validated results in plain language, knowing the figure they get back is the qashqade number. That is a powerful combination, and I am proud of what the team has built.”
Gregor Kreuzer, CPO of qashqade, added:
“What we have done is open qashqade up through a common standard, so the AI tools our clients already use can talk to it directly. You can ask in plain language, trigger a run, and get the result back inside whatever you are working in. The important part is what we deliberately kept off-limits: the AI triggers and reads, but the calculation itself is still done by our engine, the same way it always has been. The engine didn’t change, we simply added a door. And because it is built on an open protocol, clients are never locked into one tool; if a system speaks the protocol, it can connect.
To learn more about qashqade’s MCPs, contact our team.
