tracking: current context index — milestone, critical path, what an agent may pick up

[atomchung]

Status

Active execution routing — updated 2026-07-31 after one owner QA round, against main@e58fef9; draft PR #661 currently at 50c9c4a.

This issue owns no implementation, product contract, roadmap decision or acceptance evidence. Lifecycle authority: docs/issue-lifecycle.md; product roadmap/scope authority: #475.

Daily brain and agent pickup contract

This issue is the repository's mutable current control plane. A daily repository scan updates this body with the latest executable truth and appends one dated comment recording only what changed since the previous scan.

The scan must:

1. read latest main, open and recently changed PRs, tracking: current context index — milestone, critical path, what an agent may pick up #27, the roadmap guard, and only the directly referenced owning/acceptance issues;
2. reconstruct the current milestone, accepted versus merely merged work, active owners, blockers, dependency order and safe parallel lanes;
3. keep a short Claimable now section naming only work another agent can start without an unresolved owner decision or an existing implementation owner;
4. update or reuse an existing owning issue before proposing a new one; create a new issue only for a distinct executable outcome with no current home;
5. append a dated delta comment explaining additions, removals and priority changes. The comment is history; this body remains the current truth.

An open issue, a priority prefix or a green test suite does not make work claimable. Research, Discussion, Parked, Deferred and Superseded issues are never implementation pickups unless a later maintainer disposition resolves their decision and promotes them here.

Claim protocol

• An existing open PR means the implementation issue is already claimed. A second agent may review it, but must not start a competing implementation branch unless the control plane explicitly splits ownership.
• For unclaimed work listed under Claimable now, claim it before editing by assigning the issue to the working GitHub agent/account. If that agent cannot be assigned, add CLAIM — <agent/session> — branch <name>.
• Open a draft PR linked to the owning issue as the first durable work receipt. One issue has one active implementation owner unless the issue explicitly names disjoint file ownership and merge order.
• A claim ends when its PR merges/closes or the claimant posts a release comment. Stale work is not silently duplicated.
• Implement the owning issue's accepted outcome and stop condition. Record adjacent findings; do not absorb them into the PR.

Claimable now

Exactly two implementation leaves are safe to claim, one per QA repair front:

1. Front A — [defect·P1] A currency that appears only in a cash-flow row is added to the USD total at 1.0 — #612's refusal has held currencies as its domain, and no rate is ever requested #649: widen the shared FX universe to every currency entering account aggregation, fetch it, then fail closed if still unavailable. Maintainer decision is recorded on [defect·P1] A currency that appears only in a cash-flow row is added to the USD total at 1.0 — #612's refusal has held currencies as its domain, and no rate is ever requested #649.
2. Front B — [defect·P1] add-cash refuses the session it exists for: the drift gate compares question_queue and state_snapshot whole, so the card-beat cash answer is unrecordable #665: make the declared post-card cash-answer lifecycle executable without mistaking expected cash-derived projection changes for source drift. Maintainer decision is recorded on [defect·P1] add-cash refuses the session it exists for: the drift gate compares question_queue and state_snapshot whole, so the card-beat cash answer is unrecordable #665.

Do not claim #664, #673, #667, #677 or later leaves until the preceding issue in their front has landed or explicitly released ownership.

PR #661 is not a third active front. Its current head is frozen pending the scope split recorded on the PR; no competing branch should be created.

Current milestone and acceptance truth

M1 remains owner-unaccepted.

The owner completed one real QA round, but #610 did not receive an evaluable TradeEvaluation answer:

• [defect·P1] One unmatched sell anywhere in the history disables consider for the whole account — the review lane discloses the same condition and continues #673's whole-book integrity gate refused the real consider path before consequence computation.
• [design·P1] When consider cannot compute, no contract owns the answer — so the agent narrates the gap instead of framing the decision #674's unowned refusal state then produced process/gap narration rather than useful decision framing.
• [QA·dogfood] Judgment-quality dogfood — does a real consider answer now carry a supported insight? #610 therefore records blocked, not accepted. Its four judgment-quality checks must be rerun on a newly frozen exact main SHA after the repair boundary.

This is distinct from severity and priority:

• [defect·P1] A currency that appears only in a cash-flow row is added to the USD total at 1.0 — #612's refusal has held currencies as its domain, and no rate is ever requested #649/[defect·P1] A TWD position's cost basis reaches the question surface labeled USD, and the "largest cost" ranking compares raw units across currencies #664 are truth-critical even though they had no milestone assignment.
• [defect] A light-tier weekly_review cannot produce a receipt that is both honest and verifiable — the plan says the cash anchor is not applicable, the route contract demands one anyway #677 blocks honest QA evidence, not the runtime user outcome itself.
• [QA·dogfood] Judgment-quality dogfood — does a real consider answer now carry a supported insight? #610 remains the M1 owner-live acceptance owner; [QA·M1] Owner-live TradeEvaluation acceptance #488 stays deferred/non-blocking.

Broad M2 remains closed. The previously approved #403/#450 rationale-memory lane is paused at a safe split boundary while the two QA repair fronts are active.

Current repository truth

• Latest observed main: e58fef9, merged PR fix(ledger): a position the user never sold keeps its cycle across a new declaration (closes #539, closes #536) #659.
• PR fix(ledger): a position the user never sold keeps its cycle across a new declaration (closes #539, closes #536) #659 closed [bug·design·P2] Every new snapshot declaration re-asks for theses it was already given #539/[bug·design·P2] A review-lane anchor adoption spends the user's holding-duration answer #536: a continuously held position preserves cycle start, evidence basis and cycle_seq across declarations. Its PR reports the full offline suite passing; this scan could not retrieve a combined CI status for the merge SHA, so current-main CI is not independently confirmed here.
• Open PRs: draft feat(rationale): a canonical stream for why the user says they still hold a position (refs #403, refs #450) #661 and stale-base documentation PR docs(registry): profile.md has no writer, and both surfaces that claimed one now say so (refs #58) #614.
• Draft feat(rationale): a canonical stream for why the user says they still hold a position (refs #403, refs #450) #661 current head: 50c9c4a; its PR workflow completed successfully. The PR body still claims Front-B-only scope, but the branch now includes record-rationale and next-review integration plus unresolved integration findings. It must be split before further acceptance review.
• [docs] AGENTS.md, the first-review flow and the engine disagree about which question kinds may carry a private surface #666 is a narrower QA reproduction of [defect·contract] The instruction floor says two question kinds may be re-worded; the code and its own reference say six #657 and is not a separate implementation owner.

QA repair Front A — truth and usable pre-trade evaluation

One issue and one PR at a time, in this order:

1. [defect·P1] A currency that appears only in a cash-flow row is added to the USD total at 1.0 — #612's refusal has held currencies as its domain, and no rate is ever requested #649 — cross-currency cash-flow aggregation. Fetch every currency that enters the aggregate; remove 1.0 identity fallback. False account-level figures are the highest-severity finding (the synthetic reproduction was wrong by 28× while claiming full confidence).
2. [defect·P1] A TWD position's cost basis reaches the question surface labeled USD, and the "largest cost" ranking compares raw units across currencies #664 — mixed-currency question facts and ranking. The engine supplies native amount/currency for the stem and one normalized value for selection. No agent FX arithmetic; no raw-unit ranking.
3. [defect·P1] One unmatched sell anywhere in the history disables consider for the whole account — the review lane discloses the same condition and continues #673 — one per-holding integrity warning disables all of consider. Exclude and disclose the affected holding; keep only structural/unscoped corruption fatal. A premise on the excluded holding refuses by name.
4. [design·P1] When consider cannot compute, no contract owns the answer — so the agent narrates the gap instead of framing the decision #674 — bounded framing when consequence genuinely cannot compute. Use only frozen, already-computed facts; frame user-nominated options without recommending one or creating a second consequence engine.

Stop after #673 if the repaired real route now reaches a useful answer; #674 is still needed as a contract leaf, but it must not be used to preserve a recoverable refusal.

QA repair Front B — review lifecycle and honest evidence

One issue and one PR at a time, in this order:

1. [defect·P1] add-cash refuses the session it exists for: the drift gate compares question_queue and state_snapshot whole, so the card-beat cash answer is unrecordable #665 — add-cash dead end. The answer the flow asks for must reach the amended card on the session's frozen facts.
2. [defect] planned_entry declares no answer requirement, then preview refuses without a captured thesis — a dead end reachable only by following the contract #667 — planned_entry contract dead end. Require and accept thesis capture in the same exchange; keep the honesty validator strict.
3. [defect] A light-tier weekly_review cannot produce a receipt that is both honest and verifiable — the plan says the cash anchor is not applicable, the route contract demands one anyway #677 — light-tier weekly receipt contradiction. Derive the cash-event obligation from the executed plan; not_applicable means zero event, never a fabricated decline.
4. [process·qa] Cash-anchor recompute mid-first_review has no documented receipt slot for its second preview card #663 — second preview receipt. Receipt exactly one preview artifact: the final pre-commitment card used for the rule decision, after the recompute loop settles.
5. [design·process] Cash-anchor ask has no format contract — a percentage answer needs three round-trips to resolve #662 — cash input friction. Only after the lifecycle works, choose the smallest explicit input contract; do not add percentage arithmetic without naming its denominator and putting the computation in the engine.

Runtime completion (#665/#667) precedes acceptance-instrument cleanup (#677/#663), even when the latter makes QA verification red.

Queued after the two repair fronts

These remain valid, but are not launch-blocking pickups now:

1. [bug·design·P1] The revisit backlog shows the return rate its own ranking rejected — and prices a partial trim as if the whole position had gone #670: revisit card must display the dollar-impact magnitude that actually ranked the item; partial trims must not read like whole-position outcomes.
2. [bug·gate·P1] R1 and A-13 both pass a card that violates them — each gate enumerates a symptom instead of naming its mechanism #671: fix mechanism-level R1 and CJK punctuation gates, then settle one locale rendering for percentage points.
3. [defect·P2] consider's own live-resolved prices are reported as mode=agent_feed, not engine_fetch #654: truthful consider price provenance (engine_fetch versus agent_feed). M1 provenance debt, lower harm than wrong arithmetic or a dead route.
4. [defect·copy] The realized payoff sub-line shows per-trade averages without saying so, four lines under the realized total #668: label realized payoff amounts as per-trade averages; small mechanical copy leaf.
5. [defect·contract] The instruction floor says two question kinds may be re-worded; the code and its own reference say six #657: synchronize private-surface eligibility and preserve recalled-statement grounding. [docs] AGENTS.md, the first-review flow and the engine disagree about which question kinds may carry a private surface #666 contributes evidence only.
6. [bug·a11y·P2] Ruling 8 (exactly one L1, and it is Block 4) is defeated in dark mode — the class is present, the emphasis is not #672: dark-mode/accessibility follow-up.

#675, #676, #658, #650, #446, #660 and other research/future-memory items remain deferred or serially later. Do not mix them into QA repair PRs.

PR #661 disposition

Current head may not merge as one combined slice.

1. Keep feat(rationale): a canonical stream for why the user says they still hold a position (refs #403, refs #450) #661 as the canonical stream/schema/fold/bounded-reader leaf only, including corruption fail-closed, merged-chain ordering, fsync/durability and privacy/registry proof.
2. Move direct entry and next-review integration to a separate [feat·M2] Append-only rationale refresh anchored to a position cycle #403-owned draft PR stacked after the accepted stream, carrying the already-found fixes and unresolved findings.
3. Update feat(rationale): a canonical stream for why the user says they still hold a position (refs #403, refs #450) #661's body, head evidence and test count after the split; then request an independent state-integrity review.
4. Do not add condition mapping, same-ticker consider, profile verdicts or StandingRule work.

Verification and privacy gates for every repair PR

• Owning issue acceptance criteria and the maintainer disposition are both satisfied.
• Focused regression plus full offline suite passes.
• The relevant mutation fails.
• PR CI is green and the log identifies the tested tree.
• Failure leaves no pending session, canonical write or misleading artifact.
• Public evidence uses synthetic/generic data only; real trade, holding, motive and card content stay local.
• A user-visible path needs a real delivery/receipt proof where applicable; generated artifacts alone are insufficient.

Next order

1. Claim and implement [defect·P1] A currency that appears only in a cash-flow row is added to the USD total at 1.0 — #612's refusal has held currencies as its domain, and no rate is ever requested #649 and [defect·P1] add-cash refuses the session it exists for: the drift gate compares question_queue and state_snapshot whole, so the card-beat cash answer is unrecordable #665 as the only two parallel fronts.
2. Advance serially through [defect·P1] A TWD position's cost basis reaches the question surface labeled USD, and the "largest cost" ranking compares raw units across currencies #664/[defect·P1] One unmatched sell anywhere in the history disables consider for the whole account — the review lane discloses the same condition and continues #673 and [defect] planned_entry declares no answer requirement, then preview refuses without a captured thesis — a dead end reachable only by following the contract #667/[defect] A light-tier weekly_review cannot produce a receipt that is both honest and verifiable — the plan says the cash anchor is not applicable, the route contract demands one anyway #677/[process·qa] Cash-anchor recompute mid-first_review has no documented receipt slot for its second preview card #663 respectively; do not batch each chain into one PR.
3. Split feat(rationale): a canonical stream for why the user says they still hold a position (refs #403, refs #450) #661 without expanding it; keep M2 paused.
4. Freeze a new exact main SHA and rerun [QA·dogfood] Judgment-quality dogfood — does a real consider answer now carry a supported insight? #610 after [defect·P1] One unmatched sell anywhere in the history disables consider for the whole account — the review lane discloses the same condition and continues #673 reaches main and Front B has a safe review boundary.
5. Only after that acceptance verdict choose between [design·P1] When consider cannot compute, no contract owns the answer — so the agent narrates the gap instead of framing the decision #674 completion, card-content leaves, or resuming the split [feat·M2] Append-only rationale refresh anchored to a position cycle #403 integration.

Stop conditions

• Wrong confident arithmetic outranks a clean refusal; a clean refusal outranks copy polish.
• Runtime user completion outranks receipt/tooling cleanup, but acceptance cannot be declared until the receipt is honest.
• A green suite does not override a state-integrity or user-delivery finding.
• No giant QA cleanup PR: one owning issue, one reversible leaf, one acceptance proof.
• No third active implementation front.
• Merged code is not owner-visible acceptance.

[atomchung]

2026-07-04 凌晨 loop session 盤點(多 session 對齊用)

本輪開了 7 個 PR(全 CI 綠,待 review/merge),issue 全量 41 → 已解 10 個:

PR	closes	一句話
#69	#44	MIT LICENSE + 英文 TL;DR;repo description/topics 已直接 gh 設定生效
#70	#42 #43	隱私措辭架構中立 + card-feedback.yml 表單(Stage 0 回收管道落地)
#71	#61	TR_JSON/state 契約測試(38 斷言)+ CYCLE_ID_RE 單一事實源 + SKILL 收尾片段煙霧測試
#72	#62	合成價格路徑測試(11 條:賣太早邊界/β=2/α=rf 恆等式/處方四分支)+ 每週非阻塞 network-smoke job。⚠️ stacked on #71,先 merge #71
#73	#55 #56	Step 2 強制 AskUserQuestion;Step 3 規矩用戶親選 + FINAL_RULE=SKIP 語意
#74	#47	onboarding:需要 Claude Code 明示 + PEP 668 venv 三行 + ModuleNotFoundError 補救
#75	#68	eval-design ↔ EVALS ↔ card-spec 三檔對齊 + C-5/C-6 新判準。C-2 判準依賴 #73 先/同 merge

merge 順序建議:#71 → #72;#73 → #75;其餘任意(hunks 不重疊)。

剩餘 code 隊列(下個 session 可直接接,無需討論):

• [P1·第一印象] mock_trades 頭號洞已漂移 + README demo 數字失真(NVDA 佔 92%、round-trip +1940%) #45 mock 漂移(建議走修法 b「demo 凍結價格快照」,可與 [測試·P1] 「全程離線、確定性」宣稱在本機是假的:實測連網、隨行情漂、同輪混用成本/市值基礎;dedup 語意只釘一半 #64 一起修)+ README 卡/PNG 重拍
• [測試·P1] headline 選卡是「測試自己測自己」+ severity 無錨:tier 權重翻轉、稀釋 10 倍都全綠存活 #63 [測試·P1] 「全程離線、確定性」宣稱在本機是假的:實測連網、隨行情漂、同輪混用成本/市值基礎;dedup 語意只釘一半 #64 [測試·P2] 測試工程衛生:CI 版本/依賴不鎖、_DRIVER_MAP 全域耦合、test_state_loop 混居 engine 目錄 #65(測試 P1/P2:headline 自測自/離線宣稱/工程衛生——[測試·P1] 「全程離線、確定性」宣稱在本機是假的:實測連網、隨行情漂、同輪混用成本/市值基礎;dedup 語意只釘一半 #64 已有先例:test: TR_JSON/state 契約測試 — SKILL 唯一消費介面上紅綠燈 (closes #61) #71 的 yfinance shim 可直接推廣)
• [P2·誠實] 輸入層靜默丟資料:load() 無「讀入/跳過」計數,與卡面的數字誠實不一致 #50 load() 讀入/跳過計數 · [P2·邊界] 非美股/多幣別:台股港股 CSV 會靜默算錯 — 至少明示邊界 + 偵測警告 #51 非美股偵測警告 · [P2·留存] weekly loop 缺用戶端操作指引:第二週該匯什麼 CSV?卡尾沒有回訪 hook #52 weekly loop 用戶指引 · [P2·上手] demo 模式的 Step 2 沒有特例:對假交易問用戶動機,答不出 #53 demo Step 2 特例
• [UX·P1] 卡上數字 4 處 edge case 沒守:雙 ROI 並列、∞、mock alpha、insufficient_data gate 漂移 #21 卡面數字 edge case · [UX·P3] The share-safe card is documented but undiscoverable to the user (narrowed from 5 discoverability holes) #22 discoverability · 支援同日同價的多筆獨立成交（目前會被去重誤殺） #14 同日同價去重誤殺
• [GTM·P2] Demo materials describe the step-② dialogue but never show an answer changing the card #46 差異化展示(README 加 Step 2 對話示例)· [P1·分發] 多 agent 可移植性:Codex / Antigravity 已原生支援 SKILL.md 標準,只差安裝指引 + 實測 #49 多 agent 安裝指引(要實測 Codex/Antigravity)

產品方向 bucket(留 owner 整體討論,本輪不動):
#3(Stage 0,唯一真命門)· #12(投資 OS Epic)· #16(風格偵測 research)· #31–33(已裁非 MVP)· #34–38(Phase 2 批次)· #54(舉證制,綁 eval harness B-1 落地)· #57/#58(策略庫/profile 推斷)· #60(e2e 驗收,綁 harness)· #15(做空)· #8(chat 版 PARKED)· #6 #5(P2 舊件)

—— loop session by Claude(worktree zen-hugle-eb5897)

[atomchung]

2026-07-05 開發過程盤點 → 5 個 process issue（多 session 對齊用）

對這個 repo 一個月開發史（108+ 番號）做了一輪「並行／需求不清晰／變更」三類問題盤點，缺口逐條拆成 issue（皆含證據＋解法選項＋建議，留 owner 拍板）：

issue	一句話	便宜的第一步
#110	修 bug 只修發現的實例，同根因缺全面掃描（#93→#98→#108 拆股 fixture 三連）	CLAUDE.md 加一條掃描慣例
#111	zoom-out review（跨 diff 疊加＋persona 壓測）驗證有效但只跑過一次，未固化成批次 merge 前節奏	CLAUDE.md 固化觸發點＋三 smell trigger
#112	關鍵產品假設無時效標記，過時結論被跨 session 複讀（Stage 0 話術案例，owner 2026-07-05 已糾偏）	#3＋BACKLOG.md:83 補更正註記
#113	跨檔同步靠人記（eval 三檔／README 雙語／demo 卡多處），缺機械 checker	CLAUDE.md 先列鏡像對照表
#114	worktree/branch 無清理節奏，大掃除後又堆回 11 worktree / 20+ branch	批次 merge 收尾順手清＋SOP 下沉 CLAUDE.md

盤點的第四類（agent 執行品質：工具結果須硬證據驗證、別在錯的 base 下結論）刻意沒開 issue——那是 session 工作紀律不是 repo 缺陷，已在 session 工作記錄層固化；若要下沉可併入 #110/#111 落 CLAUDE.md 時一起。

並行註記：本批開出期間，#109（knowledge/ai-errors.md 跨 repo AI 錯誤帳本 PR）由另一 session 同時段冒出——與本批無語意對撞，主題與上述第四類親緣，特此互認留痕。

—— 盤點 session by Claude（worktree infallible-feistel-973c96）

[atomchung]

2026-07-05 pm 批次收尾(merge + 大掃除)

本輪 merged(依序):

• docs(claude-md): 並行開發慣例 — 認領/掃同根因/zoom-out 節奏/鏡像對照表 (closes #111, closes #114) #118 CLAUDE.md 並行開發慣例(自動 close [process·P1] zoom-out review 固化：批次 merge 前跑跨 diff 疊加檢查 + mock persona 壓測 #111/[process·P2] worktree/branch 無清理節奏：大掃除後又堆回 11 worktree / 20+ branch #114)→ main 1016dcd
• chore(hooks): PreToolUse commit gate — block engine commits when tests red #119 PreToolUse commit gate hook → main 0d11da2
• docs(readme): 分檔雙語,英文版設為 GitHub 主入口 + 繁中移至 README.zh-TW.md #107 README 分檔雙語(英文主入口 + README.zh-TW.md)→ main 2c4161a。衝突解法=英文版補譯 fork 後 fix(demo): demo_weeks 對齊 SKILL 收尾契約(ai_pct 四鍵)+ 比例人話顯示 + README 曝光 #116/refactor(engine): drop is_demo — call is data-agnostic, demo = fake data through the real call #91/feat(skill): 試駕模式 — 沒資料也能走完四步,三防護(不落盤/標演練/卡標示範) (closes #53) #117 的 8 處變更,zh-TW 直接取 main 最新全文;CLAUDE.md 的雙語規則改寫進 docs(claude-md): 並行開發慣例 — 認領/掃同根因/zoom-out 節奏/鏡像對照表 (closes #111, closes #114) #118 的鏡像對照表行
• test: 5 個 engine 邊界情境 persona fixture(pyramid/insufficient/noisy-broker/rotator/panic-seller) #81 五組 engine 邊界情境 fixture → main 54a3abf。合併前修了 sample_rotator 的 NVDA 拆股前價格(同 [bug·P0] 拆股標的 fixture 用今日拆股後價填在拆股前交易 → adjust_for_splits 雙重縮放,污染出場紀律/未實現% 訊號 #93/fix(mock): mock_trades NVDA 拆股前建倉補名目價,消除 adjust_for_splits 雙重縮放 #108 型,qty÷10 price×10),其餘 4 fixture 掃過無拆股窗口;README 兩檔 sample 計數 7→12 同步

大掃除(#114 的 SOP 首次執行):worktree 10→3(留 main / 本 session / 新並行 session xenodochial-clarke-36abde);本地 branch 34→5;remote branch 35→4。刪除標準=PR MERGED 或 commit 可達 main,逐一驗證後才刪。

刻意保留:

• remote claude/beautiful-torvalds-44cb09:無 PR,2026-06-20 雙前端 OS PRD 存檔(docs/requirements.md),刪了會丟內容,owner 決定去留
• claude/naughty-chatelet-8853d0(test: 回歸測試套件 + 一鍵 runner + CI(防止迭代改壞引擎) #25 CLOSED)已刪,內容被現行測試套件取代;如需可從 PR test: 回歸測試套件 + 一鍵 runner + CI(防止迭代改壞引擎) #25 頁面 restore branch

當前 open PR(皆 owner 決策項):

• docs(knowledge): add ai-errors.md for cross-repo AI error ledger #109 ai-errors.md 進 public repo?(可合/可 pending)
• docs: 隱私措辭誠實化 + Step 4 反饋落地管道 (closes #42, closes #43) #70 隱私措辭誠實化(closes [P0·分發] Step 4 反饋迴路沒有落地管道 — Stage 0 收不到「戳中沒」 #42/[P0·信任] 隱私措辭過度承諾:「交易不離開你的電腦」在 LLM 架構下不成立 #43):注意 README/SKILL 在它 fork 後已被大改,合前需 rebase + 語意重驗(README 隱私段現行措辭已部分吸收其意圖)

—— 批次收尾 session by Claude(worktree sleepy-boyd-6cae2f)

[atomchung]

Context-index delta — 2026-07-29 21:xx Asia/Taipei

The body predates the latest merge sequence and should not be read as the exact current SHA/PR map.

• Latest main: b035d159 (fix(data): every source records the book at its own time, and completeness stops disqualifying one (closes #549) #562).
• Open PRs: none at this check.
• fix(data): every source records the book at its own time, and completeness stops disqualifying one (closes #549) #562 head CI: tests completed successfully.
• Current product milestone remains M0, unaccepted.
• New next agent-executable leaf before formal [QA·M0] Owner-live current-book and delivery acceptance on Claude and Codex #486: reopened [bug·contract·M1-prep] Host entry points still teach retired incomplete-snapshot semantics after #562 #549, limited to synchronizing AGENTS.md / SKILL.md with fix(data): every source records the book at its own time, and completeness stops disqualifying one (closes #549) #562's canonical recorded-book semantics and adding a drift test. [QA·M0] Owner-live current-book and delivery acceptance on Claude and Codex #486 remains the owner-live exit gate after that correction.
• M1 status changed from discovery-only: feat(safety): a two-sided answer is checked against the result it claims to describe (refs #414) #547/test(eval): the trigger matrix gets a corpus and a scorer, and nothing that can call a host (refs #458) #548/refactor: the stored artifact is a TradeEvaluation, in every place it is named (refs #479) #551/feat(consider): what the user said gets frozen beside what the engine computed (refs #479) #555 landed the independent provenance validator, trigger corpus/scorer, TradeEvaluation vocabulary and DecisionContext/persistence foundation. Overall M1 remains unintegrated, release-unverified and UX-unaccepted; [feat·M1] Bounded TradeEvaluation with visible two-sided challenge #479 Wave B and [QA·M1] Owner-live TradeEvaluation acceptance #488 still own those gates.
• Closed component states on [safety·M1] Production provenance gate for decision-surface answers #414/[eval·P1] Implicit trigger reliability after explicit-entry M1 #458 must not be read as proof that production validator wiring or host/locale release evidence has passed; [QA·M1] Owner-live TradeEvaluation acceptance #488's explicit preconditions still govern.

Detailed M1 delta is recorded on #479; the host-contract blocker is recorded on #549 and promoted on #486.