I Built an AI Money Agent That's Structurally Incapable of Touching the Money

Yaseen Khatib, a senior full-stack AI engineer, built Sable, a local-first personal finance agent that stores all data on-device in SQLite and uses OpenAI function calling only to generate proposals rather than execute transactions. The agent's model output renders as a Review & Confirm card, so a hallucination can at worst produce a dismissible card and never a wrong ledger entry. Khatib argues the propose/confirm boundary and local-context pattern apply directly to enterprise agent deployments handling sensitive data like healthcare records and internal financials.

EXECUTIVE TEARDOWN // TL;DR Personal-finance apps ask for the most sensitive data a person owns and then ship it to someone else's cloud. AI finance apps go further: they hand that data to a language model and let probabilistic text decide what happens to real money. I built Sable to reject both premises at the architecture level — a local-first AI financial agent where all data lives on-device in SQLite, and where the model can propose but is structurally incapable of committing . Sable is a React Native app with no cloud backend. Every debt, every payment, every balance lives in on-device SQLite — full stop. When the AI layer needs context "how is my spending pacing this month?" , it queries the local database. What crosses the network to the model is a distilled, minimal context — never the ledger. Most products bolt privacy on as a policy. Sable has it as a topology: there is no server to breach because there is no server. The agent uses OpenAI function calling — but every function call is a dry run . When the model decides "log a ₹5,000 payment against the car loan," that intent renders as a Review & Confirm card in the UI. The model's output is a proposal object; the database mutation only executes when a human taps confirm. An LLM hallucination in Sable can produce, at worst, a card you dismiss. It can never produce a wrong number in your ledger. The question that should govern every agentic product: what is the blast radius of the model's worst output? In Sable the answer is "one dismissible card" — by architecture, not by prompt engineering. Every enterprise deploying agents faces Sable's problem in costume: healthcare records, legal documents, internal financials — data that wants AI leverage but cannot tolerate AI authority. The propose/confirm boundary and the local-context pattern transfer directly: give the model read access to distilled context, render its intents as reviewable artifacts, and reserve the commit for a human or a deterministic policy. I built the reference implementation into a product I use every day — the full breakdown is on Sable's product page https://yaseenkhatib.streamerosai.com/products/sable/ . YK Yaseen Khatib · MERN + AI Architect Ships autonomous AI products solo — five in the last twelve months. More about Yaseen → https://yaseenkhatib.streamerosai.com/about/ I'm Yaseen Khatib — a Senior Full-Stack AI Engineer MERN + TypeScript who ships production AI systems solo. Open to senior and lead roles, remote or on-site. Get in touch → https://yaseenkhatib.streamerosai.com/ contact See what I've shipped https://yaseenkhatib.streamerosai.com/products/ Originally published at yaseenkhatib.streamerosai.com/blog/sable-ai-agent-never-touches-money/ https://yaseenkhatib.streamerosai.com/blog/sable-ai-agent-never-touches-money/ .