Checking if financial processes can be bypassed before deployment
State-machine verification catches process logic flaws before deployment, but BPMS market is mature.

TLA+ state machines with LLM syntax generation — formal methods for non-engineers.
Teams designing systems, facilitators running process discovery, engineers documenting workflows
TLA+ Toolbox · Mermaid.js · Lucidchart
I made this tool to help understand large business processes that can be modelled as a single state machine.
The core loop of this is to enable to walk stakeholders through discussing each step, adding comments, and reiterating with an LLM of their choice to generate the TLA+ syntax on the left.
Users can click through the green state nodes to see how things work visually.
You can see some sample state machines in the dropdown in the top left.
The power would come from getting folks on the same page of the business process, so engineers can think about how best to automate/modernize the system once they get all of the context.
Additionally, once you have a single canonical model of the process, you could in theory send into a dark factory to produce the software as quickly as possible, but now we can allocate more time in the entire software lifecycle at the design and alignment before coding now that coding is faster.
Additional work could be done using this book for guidance, Workflow Modeling: Tools for Process Improvement and Application Development, https://www.amazon.com/Workflow-Modeling-Improvement-Applica...
MIT licensed repo: https://github.com/RCSnyder/tlaplus-process-studio
Let me know what you think!
Thanks!
State-machine verification catches process logic flaws before deployment, but BPMS market is mature.
TLA+ model checking without learning TLA+ — build fails if spec and code diverge.
PHP payment domain model handling state machines and multi-provider abstraction better than Omnipay.
HEXACO psychometrics plus dual-process chat is ambitious avatar infrastructure.
Runs Thinking Machines-style voice agent on a laptop CPU with no GPU required.
Stops stateful AI agents from OOMing using Linux Pdeathsig and socket-based cleanup.