Thriving Quest 02 · Public build
I don't want money living rent-free in my head. For me to feel fully confident in our resource planning, I need a regular and robust process for historical reporting, short-term budgeting, long-term financial planning, and tracking progress toward our financial goals.
This quest adds an extra layer: a key stakeholder, my wife. I'll manage the automation and workflows, but we make financial decisions together, so every output needs to be easy and secure for both of us to access on the web.
This thriving quest is to move from haphazard financial archaeology to a proactive and regular family finances routine: owned source data, trustworthy automation, recurring planning rituals, clear reporting, celebrations as we reach our financial goals, and specific actions tied to the life we're building.
Current problem
Irregular attention may be slowing our progress toward our family's financial goals. A lack of clarity and precision may also be shaping suboptimal financial behaviors without us realizing it.
Thesis
Agents can help us achieve financial freedom faster and with less stress.
Current progress
The ledger is 99% categorized, a September-through-December forecast runs from expected transactions and matched a hand-built rollup to the cent, and my wife and I committed our first monthly budget against it. The savings picture and the review surface still need work.
Follow along
Subscribe to the RSS means updates come to you - no account, no algorithm. Paste this into any feed reader:. The first essay will graduate from the build notes once the stable loop has real family mileage.
“The ability to do what you want, when you want, with who you want, for as long as you want, is priceless.”
I want our financial system to be penny-perfect and instantly accessible.
As end users, my wife and I have both regularly scheduled and ad hoc needs. On a regular cadence, we review budgets, assess goal progress, and plan our expected financial behaviors into the foreseeable future. At unpredictable moments, we also need to conversationally assess everything from this week's budget and spend-anomaly alerts to net-worth modeling for retirement. Anything related to money should feel easy and intuitive.
Underneath that experience, the data stays secure, scheduled loops are hardened, and the system runs proactively, nudging us toward better planning and saving.
Our ultimate goal is financial freedom. Getting there requires a better way to track and manage our family finances: one that makes progress visible, helps us act sooner, and lets us celebrate milestones along the way.
The operating loop now exists. Since the July audit, the ledger reached 99% categorization, fourteen expected-transaction lines grew into a dated forecast, and the forecast, budget, and compare loop went live across the monthly review, the quest home, and the forecast page. The first forecast covered September through December and matched the hand-built rollup to the cent. My wife and I committed the first monthly budget against it. October is the first month where drift and discipline can be measured, so the quest is now about proving the routine, not building it. It's still a little too manual, and I need to land the surfaces for us to collaborate. I'm overall pleased with the progress and, most importantly, it's already impacting our financial decisions.
The first design set the core rules: local-first data, full provenance, human-earned trust, and automation that increases only after demonstrated accuracy.
Statement reconciliation confirmed that the convenient feed was incomplete. Monthly statements became the verification layer instead of a file cabinet nobody opens.
Transactions from multiple institutions landed in a local SQLite database with source provenance, categorization, and tax metadata.
Credentials stay outside model context, statements are masked before parsing, and any shared web surface must be authenticated and least-privilege.
Goal tiers, watched spending categories, balance snapshots, health metrics, and a partner-friendly monthly review were wired together.
Statements, Plaid, and Tiller became independent raw layers feeding one rebuildable transaction table with explicit source priority.
A full audit separated the real working foundation from stale status claims, brittle merges, missing operating cadence, and security work still needed before shared access.
Changelog entry →The transaction ledger reached 99% categorization, July's review corrections were actioned, and fourteen expected-transaction lines began turning recurring, periodic, and one-off costs into a traceable family forecast.
Changelog entry →A September-through-December forecast matched the hand-built rollup to the cent, and the first monthly budget was committed by both of us with a written rationale for every deliberate deviation.
Changelog entry →Freshness, reconciliation, categorization coverage, and source conflicts become explicit health gates. A rebuild must reproduce the same totals without manual gold-layer patches.
Reporting, budgeting, long-term planning, and goal tracking each run on the cadence they require for three months, with decisions and commitments carried forward.
The system proposes specific transfers or spending adjustments with traceable evidence. Nothing moves without explicit human approval.
A September-through-December forecast now runs from the expected-transaction lines and matched my hand-built rollup to the cent. My wife and I committed our first monthly budget against it, with a written reason for every line we chose to change.
The ledger reached 99% categorization, July's review corrections landed, and fourteen expected-transaction lines turned known recurring, periodic, and one-off costs into a forecast I can inspect from the total down to the individual payment.
The July audit found a substantial build: 5,862 synthesized transactions and 160 categorization rules, but no trustworthy operating loop. A failed derived-layer repair proved that fixes must live in durable rules.
Financial freedom, resilience, lifestyle boundaries, and redesign triggers were written down, then translated into goal tiers, income assumptions, budget thresholds, and allocation rules.
Statements, Plaid, and Tiller became independent raw inputs feeding one synthesized ledger. Vendors could now be replaced without surrendering the family record or its provenance.
Goals, watched spending, live metrics, balance snapshots, and commitments came together in one review surface. The breakthrough was treating family finance as a recurring decision ritual, not another dashboard.
None yet - mature quest-log threads graduate here.