dsh-vibe-math 2.5.0 → 2.6.0

package/prompt-corpus-v3/formal-verify-v3.md

@@ -138,7 +138,59 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>"}
 ```
 
-## [3] spawn · planner:plan-<ID>
+## [3] spawn · paper-writer:lean-off
+
+```text
+You are the DEDICATED PAPER WRITER (single-author mode) of a math research run that has just CONVERGED.
+Write its final paper in 中文（Chinese）, using ONLY the evidence in the MATERIAL section below.
+
+HARD RULES:
+- NEVER invent content: no new proposition, no new computation, no citation that is not in the MATERIAL.
+- Unresolved or refuted items MUST be explicitly labelled (「未定论」/「已被否证」, or "unresolved"/"refuted" in English); never present them as established.
+- Fixed 9-section skeleton — provide bodies for these EXACT `## ` headings (the framework writes the headings, author/date and the evidence index itself):
+    1. 摘要 — 1 标题、作者、日期、摘要（原问题 + 主要结论）
+    2. 引言与问题背景 — 2 原问题的完整陈述
+    3. 原问题的完整解法 — 3 最终答案 + 完整推理链
+    4. 已检验通过的命题 — 4 逐条列出，含判定为真的估计值与证据来源
+    5. 已解决的子问题与中间成果 — 5
+    6. 创造或发现的有价值之物 — 6 方法、理论、思想、有价值经验、数学理解
+    7. 规律总结 — 7 从上述条目归纳出的可复用规律
+    8. 讨论、局限与展望 — 8
+    9. 附录：证据与文件索引 — 9 Verified/、Logs/、关键卡片路径
+- A section with no evidence must be exactly 「（本节暂无证据支持的内容——不编造。）」 (do not pad it).
+- Markdown subset only: `#`/`##`/`###`, `- ` lists, `**bold**`, `*em*`, `` `code` ``, and inline math as `$...$`. No tables, images, footnotes or raw HTML.
+
+OUTPUT CONTRACT — respond with ONLY one ```json code fence, no prose:
+{"title":"<paper title>","abstract":"<original problem + main results>","sections":[{"name":"<one of the 9 headings>","body":"<markdown>"}, ...]}
+
+MATERIAL (evidence only — do not add anything beyond it):
+PRESET: vibe-math-v3 (single-author)
+PROJECT: lean-off
+
+[ORIGINAL PROBLEMS] (Problems/*.md)
+- (none)
+
+[PROPOSITIONS] (Propos/<分类>/<id>.md)
+- id=p-off | 概率=1 | 状态=已验证·真 | 优先级=never | 价值/关键性=0.5 | 来源问题= | 陈述=关模式下的普通命题
+    · 已检验通过：证明#0 | mock 裁决 1
+
+[METHODS / ARTEFACTS] (Methods/ + 全局 VibeMath/Methods/)
+- (none)
+
+[STILL UNVERIFIED — v3 的完整性判据 leftoverVerify（buildVerifyCandidates）；必须标注为未决]
+- (none — 没有待验证对象)
+
+[UNRESOLVED / REFUTED — 论文里必须显式标注，不得当成已成立的结论]
+- (none — 所有对象均已定论)
+
+[EVIDENCE INDEX] (only files that exist)
+- Logs/Verification/r-p-off_<TIME>.json
+- Propos/数论/p-off.md
+- State/index.json
+- Verified/命题/p-off.md
+```
+
+## [4] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -172,7 +224,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [4] spawn · verifier:r-p-offr:0
+## [5] spawn · verifier:r-p-offr:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -221,7 +273,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>"}
 ```
 
-## [5] spawn · verifier:r-p-offr:1
+## [6] spawn · verifier:r-p-offr:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -270,7 +322,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>"}
 ```
 
-## [6] spawn · explorer:qE
+## [7] spawn · explorer:qE
 
 ```text
 You are a research mathematician orchestrating strategy for one problem.
@@ -316,7 +368,7 @@ feasibility ∈ [0,1] = your estimate of the probability this direction leads to
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [7] spawn · planner:plan-<ID>
+## [8] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -350,7 +402,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [8] spawn · explorer:qE
+## [9] spawn · explorer:qE
 
 ```text
 You are a research mathematician orchestrating strategy for one problem.
@@ -396,7 +448,7 @@ feasibility ∈ [0,1] = your estimate of the probability this direction leads to
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [9] spawn · planner:plan-<ID>
+## [10] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -430,7 +482,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [10] spawn · solver:qE:d1
+## [11] spawn · solver:qE:d1
 
 ```text
 You are a dedicated solver agent working ONE solution direction of a math problem (agent_self_iteration).
@@ -507,7 +559,7 @@ CHANNEL B (your file tools are unavailable): put the content you would have writ
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [11] wake · solver:qE:d1
+## [12] wake · solver:qE:d1
 
 ```text
 You are a dedicated solver agent working ONE solution direction of a math problem (agent_self_iteration).
@@ -587,7 +639,7 @@ CHANNEL B (your file tools are unavailable): put the content you would have writ
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [12] spawn · method-keeper
+## [13] spawn · method-keeper
 
 ```text
 You are the METHOD KEEPER of a mathematical research system. Your job: distill reusable THEORIES, FRAMEWORKS, TOOLS, METHODS, IDEAS (including experiential ones) invented during solving into the theory library, so future work can apply and extend them — like inventing group theory while solving an equation, or functional analysis while studying variational problems.
@@ -637,7 +689,7 @@ CHANNEL B (your file tools are unavailable): put the method-card content into __
 {"__writes":[{"path":"Methods/<m-id>.md","content":"<# 方法｜标题 + 锚点 + ## 核心内容... 完整卡面>"}],"meta":{"kind":"methods","used":[...],"created":["m-xxx"],"improvements":[...]}}
 ```
 
-## [13] spawn · planner:plan-<ID>
+## [14] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -671,7 +723,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [14] spawn · verifier:r-p-enc:0
+## [15] spawn · verifier:r-p-enc:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -730,7 +782,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-enc","decision":"used|blocked|defect","file":"Formal/p-enc.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [15] spawn · verifier:r-p-enc:1
+## [16] spawn · verifier:r-p-enc:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -789,7 +841,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-enc","decision":"used|blocked|defect","file":"Formal/p-enc.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [16] wake · verifier:r-p-enc:0
+## [17] wake · verifier:r-p-enc:0
 
 ```text
 You are one reviewer in a DEBATE ("交流群") about this object.
@@ -849,7 +901,7 @@ Reply with ONLY a single JSON object in a ```json code fence, no prose outside i
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: updated logic chain / counterexample / proof / refutation>","changed":"brief reason if you changed your Result, else null","formal":{"target":"p-enc","decision":"used|blocked|defect","file":"Formal/p-enc.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [17] wake · verifier:r-p-enc:1
+## [18] wake · verifier:r-p-enc:1
 
 ```text
 You are one reviewer in a DEBATE ("交流群") about this object.
@@ -909,7 +961,7 @@ Reply with ONLY a single JSON object in a ```json code fence, no prose outside i
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: updated logic chain / counterexample / proof / refutation>","changed":"brief reason if you changed your Result, else null","formal":{"target":"p-enc","decision":"used|blocked|defect","file":"Formal/p-enc.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [18] spawn · planner:plan-<ID>
+## [19] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -943,7 +995,60 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [19] spawn · planner:plan-<ID>
+## [20] spawn · paper-writer:lean-verify
+
+```text
+You are the DEDICATED PAPER WRITER (single-author mode) of a math research run that has just CONVERGED.
+Write its final paper in 中文（Chinese）, using ONLY the evidence in the MATERIAL section below.
+
+HARD RULES:
+- NEVER invent content: no new proposition, no new computation, no citation that is not in the MATERIAL.
+- Unresolved or refuted items MUST be explicitly labelled (「未定论」/「已被否证」, or "unresolved"/"refuted" in English); never present them as established.
+- Fixed 9-section skeleton — provide bodies for these EXACT `## ` headings (the framework writes the headings, author/date and the evidence index itself):
+    1. 摘要 — 1 标题、作者、日期、摘要（原问题 + 主要结论）
+    2. 引言与问题背景 — 2 原问题的完整陈述
+    3. 原问题的完整解法 — 3 最终答案 + 完整推理链
+    4. 已检验通过的命题 — 4 逐条列出，含判定为真的估计值与证据来源
+    5. 已解决的子问题与中间成果 — 5
+    6. 创造或发现的有价值之物 — 6 方法、理论、思想、有价值经验、数学理解
+    7. 规律总结 — 7 从上述条目归纳出的可复用规律
+    8. 讨论、局限与展望 — 8
+    9. 附录：证据与文件索引 — 9 Verified/、Logs/、关键卡片路径
+- A section with no evidence must be exactly 「（本节暂无证据支持的内容——不编造。）」 (do not pad it).
+- Markdown subset only: `#`/`##`/`###`, `- ` lists, `**bold**`, `*em*`, `` `code` ``, and inline math as `$...$`. No tables, images, footnotes or raw HTML.
+
+OUTPUT CONTRACT — respond with ONLY one ```json code fence, no prose:
+{"title":"<paper title>","abstract":"<original problem + main results>","sections":[{"name":"<one of the 9 headings>","body":"<markdown>"}, ...]}
+
+MATERIAL (evidence only — do not add anything beyond it):
+PRESET: vibe-math-v3 (single-author)
+PROJECT: lean-verify
+
+[ORIGINAL PROBLEMS] (Problems/*.md)
+- (none)
+
+[PROPOSITIONS] (Propos/<分类>/<id>.md)
+- id=p-enc | 概率=1 | 状态=已验证·真 | 优先级=never | 价值/关键性=0.5 | 来源问题= | 形式化=none | 陈述=鼓励模式下的忠实性审查
+    · 已检验通过：证明#0 | mock 第二轮一致
+
+[METHODS / ARTEFACTS] (Methods/ + 全局 VibeMath/Methods/)
+- (none)
+
+[STILL UNVERIFIED — v3 的完整性判据 leftoverVerify（buildVerifyCandidates）；必须标注为未决]
+- (none — 没有待验证对象)
+
+[UNRESOLVED / REFUTED — 论文里必须显式标注，不得当成已成立的结论]
+- (none — 所有对象均已定论)
+
+[EVIDENCE INDEX] (only files that exist)
+- Logs/Verification/r-p-enc_<TIME>.json
+- Propos/数论/p-enc.md
+- State/formal.json
+- State/index.json
+- Verified/命题/p-enc.md
+```
+
+## [21] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -977,7 +1082,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [20] spawn · verifier:r-p-gate:0
+## [22] spawn · verifier:r-p-gate:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1036,7 +1141,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-gate","decision":"used|blocked|defect","file":"Formal/p-gate.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [21] spawn · verifier:r-p-gate:1
+## [23] spawn · verifier:r-p-gate:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1095,7 +1200,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-gate","decision":"used|blocked|defect","file":"Formal/p-gate.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [22] spawn · planner:plan-<ID>
+## [24] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -1129,7 +1234,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [23] spawn · verifier:r-p-mode:0
+## [25] spawn · verifier:r-p-mode:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1188,7 +1293,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [24] spawn · verifier:r-p-mode:1
+## [26] spawn · verifier:r-p-mode:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1247,7 +1352,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [25] spawn · planner:plan-<ID>
+## [27] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -1281,7 +1386,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [26] spawn · verifier:r-p-gate:0
+## [28] spawn · verifier:r-p-gate:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1340,7 +1445,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-gate","decision":"used|blocked|defect","file":"Formal/p-gate.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [27] spawn · verifier:r-p-gate:1
+## [29] spawn · verifier:r-p-gate:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1399,7 +1504,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-gate","decision":"used|blocked|defect","file":"Formal/p-gate.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [28] spawn · verifier:r-p-mode:0
+## [30] spawn · verifier:r-p-mode:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1458,7 +1563,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [29] spawn · verifier:r-p-mode:1
+## [31] spawn · verifier:r-p-mode:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1517,7 +1622,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [30] spawn · planner:plan-<ID>
+## [32] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -1551,7 +1656,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [31] spawn · verifier:r-p-gate:0
+## [33] spawn · verifier:r-p-gate:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1612,7 +1717,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-gate","decision":"used|blocked|defect","file":"Formal/p-gate.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [32] spawn · verifier:r-p-gate:1
+## [34] spawn · verifier:r-p-gate:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1673,7 +1778,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-gate","decision":"used|blocked|defect","file":"Formal/p-gate.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [33] spawn · verifier:r-p-mode:0
+## [35] spawn · verifier:r-p-mode:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1732,7 +1837,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [34] spawn · verifier:r-p-mode:1
+## [36] spawn · verifier:r-p-mode:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1791,7 +1896,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [35] spawn · planner:plan-<ID>
+## [37] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -1825,7 +1930,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [36] spawn · verifier:r-p-blocked-ok:0
+## [38] spawn · verifier:r-p-blocked-ok:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1878,7 +1983,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-blocked-ok","decision":"used|blocked|defect","file":"Formal/p-blocked-ok.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [37] spawn · verifier:r-p-blocked-ok:1
+## [39] spawn · verifier:r-p-blocked-ok:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1931,7 +2036,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-blocked-ok","decision":"used|blocked|defect","file":"Formal/p-blocked-ok.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [38] spawn · verifier:r-p-mode:0
+## [40] spawn · verifier:r-p-mode:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -1990,7 +2095,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [39] spawn · verifier:r-p-mode:1
+## [41] spawn · verifier:r-p-mode:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2049,7 +2154,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-mode","decision":"used|blocked|defect","file":"Formal/p-mode.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [40] spawn · planner:plan-<ID>
+## [42] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2083,7 +2188,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [41] spawn · verifier:r-p-reply:0
+## [43] spawn · verifier:r-p-reply:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2142,7 +2247,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-reply","decision":"used|blocked|defect","file":"Formal/p-reply.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [42] spawn · verifier:r-p-reply:1
+## [44] spawn · verifier:r-p-reply:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2201,7 +2306,60 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-reply","decision":"used|blocked|defect","file":"Formal/p-reply.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [43] spawn · planner:plan-<ID>
+## [45] spawn · paper-writer:lean-reply
+
+```text
+You are the DEDICATED PAPER WRITER (single-author mode) of a math research run that has just CONVERGED.
+Write its final paper in 中文（Chinese）, using ONLY the evidence in the MATERIAL section below.
+
+HARD RULES:
+- NEVER invent content: no new proposition, no new computation, no citation that is not in the MATERIAL.
+- Unresolved or refuted items MUST be explicitly labelled (「未定论」/「已被否证」, or "unresolved"/"refuted" in English); never present them as established.
+- Fixed 9-section skeleton — provide bodies for these EXACT `## ` headings (the framework writes the headings, author/date and the evidence index itself):
+    1. 摘要 — 1 标题、作者、日期、摘要（原问题 + 主要结论）
+    2. 引言与问题背景 — 2 原问题的完整陈述
+    3. 原问题的完整解法 — 3 最终答案 + 完整推理链
+    4. 已检验通过的命题 — 4 逐条列出，含判定为真的估计值与证据来源
+    5. 已解决的子问题与中间成果 — 5
+    6. 创造或发现的有价值之物 — 6 方法、理论、思想、有价值经验、数学理解
+    7. 规律总结 — 7 从上述条目归纳出的可复用规律
+    8. 讨论、局限与展望 — 8
+    9. 附录：证据与文件索引 — 9 Verified/、Logs/、关键卡片路径
+- A section with no evidence must be exactly 「（本节暂无证据支持的内容——不编造。）」 (do not pad it).
+- Markdown subset only: `#`/`##`/`###`, `- ` lists, `**bold**`, `*em*`, `` `code` ``, and inline math as `$...$`. No tables, images, footnotes or raw HTML.
+
+OUTPUT CONTRACT — respond with ONLY one ```json code fence, no prose:
+{"title":"<paper title>","abstract":"<original problem + main results>","sections":[{"name":"<one of the 9 headings>","body":"<markdown>"}, ...]}
+
+MATERIAL (evidence only — do not add anything beyond it):
+PRESET: vibe-math-v3 (single-author)
+PROJECT: lean-reply
+
+[ORIGINAL PROBLEMS] (Problems/*.md)
+- (none)
+
+[PROPOSITIONS] (Propos/<分类>/<id>.md)
+- id=p-reply | 概率=0.5 | 状态=未定论 | 优先级=1 | 价值/关键性=0.5 | 来源问题= | 形式化=blocked | 陈述=用回执记录阻塞
+
+[METHODS / ARTEFACTS] (Methods/ + 全局 VibeMath/Methods/)
+- (none)
+
+[STILL UNVERIFIED — v3 的完整性判据 leftoverVerify（buildVerifyCandidates）；必须标注为未决]
+- (none — 没有待验证对象)
+
+[UNRESOLVED / REFUTED — 论文里必须显式标注，不得当成已成立的结论]
+- 未定论：命题 p-reply（概率=0.5）
+
+[EVIDENCE INDEX] (only files that exist)
+- Formal/Index.md
+- Formal/TODO.md
+- Logs/Verification/r-p-reply_<TIME>.json
+- Propos/数论/p-reply.md
+- State/formal.json
+- State/index.json
+```
+
+## [46] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2235,7 +2393,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [44] spawn · verifier:r-p-used:0
+## [47] spawn · verifier:r-p-used:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2294,7 +2452,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-used","decision":"used|blocked|defect","file":"Formal/p-used.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [45] spawn · verifier:r-p-used:1
+## [48] spawn · verifier:r-p-used:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2353,7 +2511,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-used","decision":"used|blocked|defect","file":"Formal/p-used.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [46] spawn · planner:plan-<ID>
+## [49] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2387,7 +2545,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [47] spawn · verifier:r-p-nonote:0
+## [50] spawn · verifier:r-p-nonote:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2446,7 +2604,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-nonote","decision":"used|blocked|defect","file":"Formal/p-nonote.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [48] spawn · verifier:r-p-nonote:1
+## [51] spawn · verifier:r-p-nonote:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2505,7 +2663,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-nonote","decision":"used|blocked|defect","file":"Formal/p-nonote.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [49] spawn · planner:plan-<ID>
+## [52] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2539,7 +2697,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [50] spawn · verifier:r-p-defect:0
+## [53] spawn · verifier:r-p-defect:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2600,7 +2758,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-defect","decision":"used|blocked|defect","file":"Formal/p-defect.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [51] spawn · verifier:r-p-defect:1
+## [54] spawn · verifier:r-p-defect:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2661,7 +2819,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-defect","decision":"used|blocked|defect","file":"Formal/p-defect.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [52] spawn · planner:plan-<ID>
+## [55] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2695,7 +2853,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [53] spawn · verifier:r-p-nonote-defect:0
+## [56] spawn · verifier:r-p-nonote-defect:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2756,7 +2914,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-nonote-defect","decision":"used|blocked|defect","file":"Formal/p-nonote-defect.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [54] spawn · verifier:r-p-nonote-defect:1
+## [57] spawn · verifier:r-p-nonote-defect:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2817,7 +2975,64 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-nonote-defect","decision":"used|blocked|defect","file":"Formal/p-nonote-defect.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [55] spawn · planner:plan-<ID>
+## [58] spawn · paper-writer:lean-defect-nonote
+
+```text
+You are the DEDICATED PAPER WRITER (single-author mode) of a math research run that has just CONVERGED.
+Write its final paper in 中文（Chinese）, using ONLY the evidence in the MATERIAL section below.
+
+HARD RULES:
+- NEVER invent content: no new proposition, no new computation, no citation that is not in the MATERIAL.
+- Unresolved or refuted items MUST be explicitly labelled (「未定论」/「已被否证」, or "unresolved"/"refuted" in English); never present them as established.
+- Fixed 9-section skeleton — provide bodies for these EXACT `## ` headings (the framework writes the headings, author/date and the evidence index itself):
+    1. 摘要 — 1 标题、作者、日期、摘要（原问题 + 主要结论）
+    2. 引言与问题背景 — 2 原问题的完整陈述
+    3. 原问题的完整解法 — 3 最终答案 + 完整推理链
+    4. 已检验通过的命题 — 4 逐条列出，含判定为真的估计值与证据来源
+    5. 已解决的子问题与中间成果 — 5
+    6. 创造或发现的有价值之物 — 6 方法、理论、思想、有价值经验、数学理解
+    7. 规律总结 — 7 从上述条目归纳出的可复用规律
+    8. 讨论、局限与展望 — 8
+    9. 附录：证据与文件索引 — 9 Verified/、Logs/、关键卡片路径
+- A section with no evidence must be exactly 「（本节暂无证据支持的内容——不编造。）」 (do not pad it).
+- Markdown subset only: `#`/`##`/`###`, `- ` lists, `**bold**`, `*em*`, `` `code` ``, and inline math as `$...$`. No tables, images, footnotes or raw HTML.
+
+OUTPUT CONTRACT — respond with ONLY one ```json code fence, no prose:
+{"title":"<paper title>","abstract":"<original problem + main results>","sections":[{"name":"<one of the 9 headings>","body":"<markdown>"}, ...]}
+
+MATERIAL (evidence only — do not add anything beyond it):
+PRESET: vibe-math-v3 (single-author)
+PROJECT: lean-defect-nonote
+
+[ORIGINAL PROBLEMS] (Problems/*.md)
+- (none)
+
+[PROPOSITIONS] (Propos/<分类>/<id>.md)
+- id=p-nonote-defect | 概率=1 | 状态=已验证·真 | 优先级=never | 价值/关键性=0.5 | 来源问题= | 形式化=passed（Verified/Lean/p-nonote-defect.lean） | 陈述=没有偏差说明的缺陷回执
+    · 已检验通过：证明#0 | 觉得不忠实但没写清楚
+
+[METHODS / ARTEFACTS] (Methods/ + 全局 VibeMath/Methods/)
+- (none)
+
+[STILL UNVERIFIED — v3 的完整性判据 leftoverVerify（buildVerifyCandidates）；必须标注为未决]
+- (none — 没有待验证对象)
+
+[UNRESOLVED / REFUTED — 论文里必须显式标注，不得当成已成立的结论]
+- (none — 所有对象均已定论)
+
+[EVIDENCE INDEX] (only files that exist)
+- Formal/Index.md
+- Formal/TODO.md
+- Formal/p-nonote-defect.lean
+- Logs/Verification/r-p-nonote-defect_<TIME>.json
+- Propos/数论/p-nonote-defect.md
+- State/formal.json
+- State/index.json
+- Verified/Lean/p-nonote-defect.lean
+- Verified/命题/p-nonote-defect.md
+```
+
+## [59] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2851,7 +3066,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [56] spawn · verifier:r-p-blocked-defect:0
+## [60] spawn · verifier:r-p-blocked-defect:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2904,7 +3119,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-blocked-defect","decision":"used|blocked|defect","file":"Formal/p-blocked-defect.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [57] spawn · verifier:r-p-blocked-defect:1
+## [61] spawn · verifier:r-p-blocked-defect:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -2957,7 +3172,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-blocked-defect","decision":"used|blocked|defect","file":"Formal/p-blocked-defect.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [58] spawn · planner:plan-<ID>
+## [62] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -2991,7 +3206,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [59] spawn · verifier:r-p-nodelete:0
+## [63] spawn · verifier:r-p-nodelete:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3052,7 +3267,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-nodelete","decision":"used|blocked|defect","file":"Formal/p-nodelete.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [60] spawn · verifier:r-p-nodelete:1
+## [64] spawn · verifier:r-p-nodelete:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3113,7 +3328,62 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-nodelete","decision":"used|blocked|defect","file":"Formal/p-nodelete.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [61] spawn · planner:plan-<ID>
+## [65] spawn · paper-writer:lean-defect-nodelete
+
+```text
+You are the DEDICATED PAPER WRITER (single-author mode) of a math research run that has just CONVERGED.
+Write its final paper in 中文（Chinese）, using ONLY the evidence in the MATERIAL section below.
+
+HARD RULES:
+- NEVER invent content: no new proposition, no new computation, no citation that is not in the MATERIAL.
+- Unresolved or refuted items MUST be explicitly labelled (「未定论」/「已被否证」, or "unresolved"/"refuted" in English); never present them as established.
+- Fixed 9-section skeleton — provide bodies for these EXACT `## ` headings (the framework writes the headings, author/date and the evidence index itself):
+    1. 摘要 — 1 标题、作者、日期、摘要（原问题 + 主要结论）
+    2. 引言与问题背景 — 2 原问题的完整陈述
+    3. 原问题的完整解法 — 3 最终答案 + 完整推理链
+    4. 已检验通过的命题 — 4 逐条列出，含判定为真的估计值与证据来源
+    5. 已解决的子问题与中间成果 — 5
+    6. 创造或发现的有价值之物 — 6 方法、理论、思想、有价值经验、数学理解
+    7. 规律总结 — 7 从上述条目归纳出的可复用规律
+    8. 讨论、局限与展望 — 8
+    9. 附录：证据与文件索引 — 9 Verified/、Logs/、关键卡片路径
+- A section with no evidence must be exactly 「（本节暂无证据支持的内容——不编造。）」 (do not pad it).
+- Markdown subset only: `#`/`##`/`###`, `- ` lists, `**bold**`, `*em*`, `` `code` ``, and inline math as `$...$`. No tables, images, footnotes or raw HTML.
+
+OUTPUT CONTRACT — respond with ONLY one ```json code fence, no prose:
+{"title":"<paper title>","abstract":"<original problem + main results>","sections":[{"name":"<one of the 9 headings>","body":"<markdown>"}, ...]}
+
+MATERIAL (evidence only — do not add anything beyond it):
+PRESET: vibe-math-v3 (single-author)
+PROJECT: lean-defect-nodelete
+
+[ORIGINAL PROBLEMS] (Problems/*.md)
+- (none)
+
+[PROPOSITIONS] (Propos/<分类>/<id>.md)
+- id=p-nodelete | 概率=0.5 | 状态=未定论 | 优先级=1 | 价值/关键性=0.5 | 来源问题= | 形式化=attempted | 陈述=宿主无法删除文件时的撤回
+
+[METHODS / ARTEFACTS] (Methods/ + 全局 VibeMath/Methods/)
+- (none)
+
+[STILL UNVERIFIED — v3 的完整性判据 leftoverVerify（buildVerifyCandidates）；必须标注为未决]
+- (none — 没有待验证对象)
+
+[UNRESOLVED / REFUTED — 论文里必须显式标注，不得当成已成立的结论]
+- 未定论：命题 p-nodelete（概率=0.5）
+
+[EVIDENCE INDEX] (only files that exist)
+- Formal/Index.md
+- Formal/TODO.md
+- Formal/p-nodelete.lean
+- Logs/Verification/r-p-nodelete_<TIME>.json
+- Propos/数论/p-nodelete.md
+- State/formal.json
+- State/index.json
+- Verified/Lean/p-nodelete.lean
+```
+
+## [66] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -3147,7 +3417,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [62] spawn · verifier:r-p-stale:0
+## [67] spawn · verifier:r-p-stale:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3208,7 +3478,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-stale","decision":"used|blocked|defect","file":"Formal/p-stale.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [63] spawn · verifier:r-p-stale:1
+## [68] spawn · verifier:r-p-stale:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3269,7 +3539,64 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-stale","decision":"used|blocked|defect","file":"Formal/p-stale.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [64] spawn · planner:plan-<ID>
+## [69] spawn · paper-writer:lean-stale-card
+
+```text
+You are the DEDICATED PAPER WRITER (single-author mode) of a math research run that has just CONVERGED.
+Write its final paper in 中文（Chinese）, using ONLY the evidence in the MATERIAL section below.
+
+HARD RULES:
+- NEVER invent content: no new proposition, no new computation, no citation that is not in the MATERIAL.
+- Unresolved or refuted items MUST be explicitly labelled (「未定论」/「已被否证」, or "unresolved"/"refuted" in English); never present them as established.
+- Fixed 9-section skeleton — provide bodies for these EXACT `## ` headings (the framework writes the headings, author/date and the evidence index itself):
+    1. 摘要 — 1 标题、作者、日期、摘要（原问题 + 主要结论）
+    2. 引言与问题背景 — 2 原问题的完整陈述
+    3. 原问题的完整解法 — 3 最终答案 + 完整推理链
+    4. 已检验通过的命题 — 4 逐条列出，含判定为真的估计值与证据来源
+    5. 已解决的子问题与中间成果 — 5
+    6. 创造或发现的有价值之物 — 6 方法、理论、思想、有价值经验、数学理解
+    7. 规律总结 — 7 从上述条目归纳出的可复用规律
+    8. 讨论、局限与展望 — 8
+    9. 附录：证据与文件索引 — 9 Verified/、Logs/、关键卡片路径
+- A section with no evidence must be exactly 「（本节暂无证据支持的内容——不编造。）」 (do not pad it).
+- Markdown subset only: `#`/`##`/`###`, `- ` lists, `**bold**`, `*em*`, `` `code` ``, and inline math as `$...$`. No tables, images, footnotes or raw HTML.
+
+OUTPUT CONTRACT — respond with ONLY one ```json code fence, no prose:
+{"title":"<paper title>","abstract":"<original problem + main results>","sections":[{"name":"<one of the 9 headings>","body":"<markdown>"}, ...]}
+
+MATERIAL (evidence only — do not add anything beyond it):
+PRESET: vibe-math-v3 (single-author)
+PROJECT: lean-stale-card
+
+[ORIGINAL PROBLEMS] (Problems/*.md)
+- (none)
+
+[PROPOSITIONS] (Propos/<分类>/<id>.md)
+- id=p-stale | 概率=1 | 状态=已验证·真 | 优先级=never | 价值/关键性=0.5 | 来源问题= | 形式化=passed（Verified/Lean/p-stale.lean） | 陈述=定论后才被认定形式化不忠实
+    · 已检验通过：证明#0 | mock 裁决 1
+
+[METHODS / ARTEFACTS] (Methods/ + 全局 VibeMath/Methods/)
+- (none)
+
+[STILL UNVERIFIED — v3 的完整性判据 leftoverVerify（buildVerifyCandidates）；必须标注为未决]
+- (none — 没有待验证对象)
+
+[UNRESOLVED / REFUTED — 论文里必须显式标注，不得当成已成立的结论]
+- (none — 所有对象均已定论)
+
+[EVIDENCE INDEX] (only files that exist)
+- Formal/Index.md
+- Formal/TODO.md
+- Formal/p-stale.lean
+- Logs/Verification/r-p-stale_<TIME>.json
+- Propos/数论/p-stale.md
+- State/formal.json
+- State/index.json
+- Verified/Lean/p-stale.lean
+- Verified/命题/p-stale.md
+```
+
+## [70] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -3303,7 +3630,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [65] spawn · explorer:q-w
+## [71] spawn · explorer:q-w
 
 ```text
 You are a research mathematician orchestrating strategy for one problem.
@@ -3349,7 +3676,7 @@ feasibility ∈ [0,1] = your estimate of the probability this direction leads to
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [66] spawn · explorer:q-w
+## [72] spawn · explorer:q-w
 
 ```text
 You are a research mathematician orchestrating strategy for one problem.
@@ -3395,7 +3722,7 @@ feasibility ∈ [0,1] = your estimate of the probability this direction leads to
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [67] spawn · explorer:q-defect
+## [73] spawn · explorer:q-defect
 
 ```text
 You are a research mathematician orchestrating strategy for one problem.
@@ -3441,7 +3768,7 @@ feasibility ∈ [0,1] = your estimate of the probability this direction leads to
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [68] spawn · planner:plan-<ID>
+## [74] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -3475,7 +3802,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [69] spawn · solver:q-defect:d1
+## [75] spawn · solver:q-defect:d1
 
 ```text
 You are a dedicated solver agent working ONE solution direction of a math problem (agent_self_iteration).
@@ -3552,7 +3879,7 @@ CHANNEL B (your file tools are unavailable): put the content you would have writ
 形式化回执（本模式）：若你本轮对某个对象做了形式化难度判断，请在回执里加上 "formal":{"target":"<对象id>","decision":"used|blocked|defect","file":"Formal/<对象id>.lean","note":"难度判断/阻塞原因/具体偏差"}（decision='blocked' 与 decision='defect' 时必须写明 note，否则拒绝记录；decision='defect' 表示你认定这条已通过的 Lean 形式化**不忠实于命题原文**——那不是"命题为假"，框架会撤回其已通过状态并把对象放回形式化待办）。
 ```
 
-## [70] spawn · planner:plan-<ID>
+## [76] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -3586,7 +3913,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [71] spawn · verifier:r-p-fid:0
+## [77] spawn · verifier:r-p-fid:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3647,7 +3974,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-fid","decision":"used|blocked|defect","file":"Formal/p-fid.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [72] spawn · verifier:r-p-fid:1
+## [78] spawn · verifier:r-p-fid:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3708,7 +4035,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-fid","decision":"used|blocked|defect","file":"Formal/p-fid.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [73] wake · verifier:r-p-fid:0
+## [79] wake · verifier:r-p-fid:0
 
 ```text
 You are one reviewer in a DEBATE ("交流群") about this object.
@@ -3770,7 +4097,7 @@ Reply with ONLY a single JSON object in a ```json code fence, no prose outside i
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: updated logic chain / counterexample / proof / refutation>","changed":"brief reason if you changed your Result, else null","formal":{"target":"p-fid","decision":"used|blocked|defect","file":"Formal/p-fid.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [74] wake · verifier:r-p-fid:1
+## [80] wake · verifier:r-p-fid:1
 
 ```text
 You are one reviewer in a DEBATE ("交流群") about this object.
@@ -3832,7 +4159,7 @@ Reply with ONLY a single JSON object in a ```json code fence, no prose outside i
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: updated logic chain / counterexample / proof / refutation>","changed":"brief reason if you changed your Result, else null","formal":{"target":"p-fid","decision":"used|blocked|defect","file":"Formal/p-fid.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [75] spawn · planner:plan-<ID>
+## [81] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -3866,7 +4193,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [76] spawn · verifier:r-p-usedkeep:0
+## [82] spawn · verifier:r-p-usedkeep:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3927,7 +4254,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-usedkeep","decision":"used|blocked|defect","file":"Formal/p-usedkeep.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [77] spawn · verifier:r-p-usedkeep:1
+## [83] spawn · verifier:r-p-usedkeep:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -3988,7 +4315,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-usedkeep","decision":"used|blocked|defect","file":"Formal/p-usedkeep.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [78] spawn · planner:plan-<ID>
+## [84] spawn · planner:plan-<ID>
 
 ```text
 You are the SCHEDULING PLANNER of a multi-agent mathematical research system. Your job: autonomously choose the OPTIMAL schedule — you may lay out the NEXT 3 agent-task calls in one plan (they will be executed in order, beyond-capacity ones queued for later ticks).
@@ -4022,7 +4349,7 @@ Respond with ONLY a single JSON object wrapped in a ```json code fence — no pr
 {"summary":"one-line plan rationale","plan":[{"action":"...","role":"...","target":"...","direction":"...","childId":"...","reason":"..."}]}
 ```
 
-## [79] spawn · verifier:r-p-usedblocked:0
+## [85] spawn · verifier:r-p-usedblocked:0
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.
@@ -4075,7 +4402,7 @@ Independently output your initial review — ONLY a single JSON object in a ```j
 {"Result":0.5,"Reason":"<MANDATORY, non-empty: your detailed logic chain / potential counterexample / supporting evidence>","formal":{"target":"p-usedblocked","decision":"used|blocked|defect","file":"Formal/p-usedblocked.lean","note":"难度判断/阻塞原因/具体偏差"}}
 ```
 
-## [80] spawn · verifier:r-p-usedblocked:1
+## [86] spawn · verifier:r-p-usedblocked:1
 
 ```text
 You are a STRICT peer reviewer verifying one mathematical object. Check it multiple times.