Skip to content

BACKLOG.md

Latest commit

 

History

History
123 lines (86 loc) · 7.08 KB

BACKLOG.md

File metadata and controls

123 lines (86 loc) · 7.08 KB
Raw

fomo-kernel backlog

Last refreshed: 2026-07-17. The target is a promotion-ready release on 2026-07-19.

North star (owner correction 2026-07-05, #112 — do not regress this)

The vision is an agent, but the next action is still one card striking one real person. The constraint is card quality — whether the card is useful enough and at a publishable standard — not distribution or finding users. The card has always been tested and always been shown to people; the real bottleneck is that the card is not yet good enough. The promotion release below is a vehicle for validating card quality, not a distribution goal in itself. Do not let the big vision steal the small thing that needs validating now.

P0 for the promotion release

Stable workflow and card production

Status: implemented, pending manual release gates.

  • Keep SKILL.md as a thin entry point.
  • Use the fixed prepare -> preview -> finalize lifecycle.
  • Resume interrupted sessions without refetching live data.
  • Commit one canonical immutable session and rebuild projections from it.
  • Render private and public cards deterministically.
  • Keep required questions and evidence completeness as code gates.

Release evidence: docs/release-2026-07-19.md, tests/test_review_v2.py, and the complete offline suite.

Thesis evolution and add evidence

Status: implemented in the v2 lifecycle.

  • Classify losing-position adds as planned tranche, new evidence, valuation change, price only, or skip.
  • Require a claim and source for new_evidence.
  • Store append-only thesis decision events tied to active cycle IDs.
  • Reconcile the evidence in future reviews rather than asking a generic averaging-down question again.

Weekly reconciliation and memory presentation

Status: implemented in the v2 lifecycle and private-card renderer, pending second-review dogfood and the manual card-quality gate.

  • Open a later review by reconciling the user's prior commitment rather than starting from zero.
  • Put due exit revisits into the required question queue, preserve the earlier exit reason, and frame the checkpoint against the actual swap outcome.
  • Keep cold-start historical exits in a summarized backlog instead of flooding required questions.
  • Surface recurring four-week problem counts and ask for a qualitative judgment when a chosen rule appears to have been broken again.
  • Mirror frozen market context and holding-horizon contradictions from the review plan onto the private card.
  • Leave skipped due checkpoints unresolved so they return; commit answered checkpoints and review marks through canonical session projections.

ETF policy

Status: implemented with conservative fallback.

  • Exempt only broad-market, regional, bond, and commodity allocation ETFs from single-name concentration.
  • Keep sector, thematic, and leveraged ETFs in concentration and stress diagnostics.
  • Give unknown instruments no exemption.
  • Disclose missing expense ratio and tracking error rather than assuming zero.
  • Add a live metadata source later without changing the policy contract.

English implementation and bilingual GTM

Status: implemented in this change, pending full verification.

  • Keep developer documentation and skill instructions in English.
  • Keep English and Traditional Chinese GTM artifacts synchronized as separate files.
  • Keep user-visible localized product copy in separate locale resources.
  • Prevent mixed-language implementation docs with a deterministic regression test.

Manual release gates

  • Complete one full Traditional Chinese run with an anonymized publishable CSV.
  • Complete the same flow in English.
  • Inspect the public card manually for amounts, dates, tickers, exact weights, session IDs, evidence text, and free-form narrative.
  • Demonstrate evidence-gate rejection and recovery.
  • Demonstrate broad ETF exemption versus thematic ETF concentration.
  • Complete a second review against the same local state and inspect prior-rule reconciliation, due checks with frozen-price swap framing, exit backlog, recurring-problem presentation, market context, and any horizon contradiction.

Shipped after the P0 gate scope was frozen

Initial snapshot onboarding adapter

Status: implemented as the v2.1 onboarding slice; validate separately from the 2026-07-19 CSV-focused manual gates.

  • Accept a normalized position JSON envelope produced locally from a table or screenshot through review.py.
  • Keep screenshot transcription local and the temporary normalized JSON outside the repository; there is no engine OCR or cloud-upload path.
  • Normalize the declared facts into the snapshot review card/state contract. A complete initial snapshot may establish the accounting anchor; an incomplete snapshot produces a bounded review without becoming an anchor.
  • Limit conclusions to cost or value weights, single-position risk, driver concentration, ETF structure, and data integrity.
  • Initialize inferred theses for open cycles without claiming historical motives.
  • Allow later transaction history to unlock supported history-dependent behavioral dimensions while ledger-derived current holdings remain canonical and unreconciled current-view claims fail closed.

P1

Post-release lens research and possible selection

  • Keep the existing rubric/ files as research assets; do not load them into current v2 questions or cards.
  • Select from a small verified lens set.
  • Apply style-specific divergence only where mechanical evidence supports a style axis.
  • Omit lens-specific interpretation when the selected lens has no explicit stance for that dimension.
  • Keep universal risk facts independent of the lens and never attribute them to a persona.
  • Do not duplicate lifecycle, state, schemas, or renderers.
  • Keep public wording as source-linked paraphrase, never a quotation or persona endorsement.

Snapshot reconciliation and repeated snapshots

  • Compare a second or subsequent declared snapshot with ledger-derived current holdings.
  • Present a narrow diff, fail closed on ambiguous causes, and write an explicit adjustment event only through the reconciliation contract.
  • Adopt a newer declared accounting anchor only after that reconciliation succeeds; do not let transaction-history import silently assert a fresh broker view.

ETF metadata enrichment

  • Add a maintained instrument source for classification, expense ratio, and tracking error.
  • Preserve the local override and conservative unknown fallback.
  • Cache data for offline and repeatable reviews.

Later candidates

  • Event-driven pre-trade check against the active process rule.
  • Personal lens distilled from repeated confirmed review patterns.
  • Richer source attribution for owner-only research workflows.
  • Automated GTM asset generation and publishing.
  • More behavior detectors only when they are measurable and bind to a testable rule.

Product boundaries

  • Process coaching is not security selection.
  • One card converges on one behavioral leak and at most one rule.
  • Trade data remains local.
  • A clean strengths card is valid when no costly leak is supported.
  • Real-user usefulness remains the final validation layer; passing automated tests is necessary but not sufficient.