The problem Armor Wallet solves

The idea is to promote self-custody of users’ funds. The EOAs in use today “promise” users that they are the owner of the assets stored in these wallets but it only takes you to lose access to these wallets and you have now lost all the funds. Account abstraction allows users to own their funds by replacing EOAs with smart contract wallets. These programmable smart contract wallets are flexible enough to process single and multiple transactions with different signature techniques and support the social recovery of wallets. With batching of transactions, problems like approve patterns in ERC-20 and infinite approvals can be solved.
Armor Wallet is a POC solution that aims to add protection against MEV attacks for transactions and reduce the barrier to entry of new users into the ecosystem by allowing paymasters to fund the transactions using a custom ERC20 token. The paymaster mints custom ERC20 tokens that can be used to pay for the gas of the transactions. The locally run bundler follows the ERC4337 specifications and adds the flashbot protect provider code that bundles up the userOp transactions and sends them to flashbot protect for MEV protection.
Armor Wallet looks to solve problems like:

1. Abstracting away the need for EOAs
2. Promotes usage of Smart Contract wallets
3. Protects batched transactions from MEV attacks to avoid mass invalidation
4. Allows users to use any ERC20 tokens to fund their transactions

Challenges I ran into

The Account abstraction space is very new and everybody is trying to come up with solutions to make it usable.

1. The project idea formation was tough as the feasibility of the project was in question owing to ambiguity in the current ecosystem of Account Abstraction. Extensive discussions with industry experts and mentors helped me out in that sphere.
2. Majority of the references are broken and often times figuring out how to make the references work was time-consuming.
3. Building certain features like paymaster and MEV resistance is completely out of the box or new and needs a ground-up implementation.