Blockchains are maintained by a network of participants that run algorithms designed to maintain collectively a distributed machine tolerant to Byzantine attacks. From the point of view of users, blockchains provide the illusion of centralized computers that perform trustable verifiable computations, where all computations are deterministic and the results cannot be manipulated or undone. Smart-contracts are written in a special-purpose programming language with deterministic semantics. Each transaction begins with an invocation from an external user to a smart contract. Contracts have local storage and can call other contracts, and more importantly, they store, send and receive cryptocurrency. It is essential to guarantee that contracts are correct before deployment since their code cannot be modified afterward deployment. However, the resulting ecosystem makes it very difficult to reason about program correctness, since contracts can be executed by malicious users or malicious contracts can be designed to exploit other contracts that call them. Many attacks and bugs are caused by unexpected interactions between multiple contracts, the attacked contract and unknown code that performs the exploit. Moreover, there is a very aggressive competition between different blockchains to expand their user base. Ideas are implemented fast and blockchains compete to offer and adopt new features quickly. In this paper, we propose a formal extensible playground that allows reasoning about multi-contract interactions to ultimately prove properties before features are incorporated into the real blockchain. We implemented a model of computation that models the execution platform, abstracts the internal code of each individual contract, and focuses on contract interactions. Moreover, we show how many features, existing or proposed, can be used to reason about multi-contract interactions.