You're reading this. The skill predicted it. A workflow that turns every post into a calibrated experiment—score, blind-predict, retro, evolve. The future doesn't reward effort, it rewards those who see the pattern first. 1M followers in a month — not luck, system.
Python
cheat-init: fix state-write ambiguity that caused content_form="a" bug
Jooonnna577d460d28a
diff --git a/skills/cheat-init/SKILL.md b/skills/cheat-init/SKILL.mdindex da2df92..1e0b70d 100644--- a/skills/cheat-init/SKILL.md+++ b/skills/cheat-init/SKILL.md@@ -92,8 +92,22 @@ allowed-tools: Bash(*), Read, Write, Edit, Glob, WebFetch, Skill> g) **混合**"记录到 `content_form` + `rubric_form_mismatch`。-- 选 a → `rubric_form_mismatch: false`-- 选 b/c/d/e/f/g → `rubric_form_mismatch: true`,cheat-status 持续提示"你的形态可能需要 bump 调权重"++**Q1 → `content_form` enum 映射**(**必须存 enum 值,不是字母**):++| 用户答 | `content_form` 写入值 |+|---|---|+| a | `"opinion-video"` |+| b | `"long-essay"` |+| c | `"short-text"` |+| d | `"podcast"` |+| e | `"tutorial-builder"` |+| f | `"other"` |+| g | `"mixed"` |++`rubric_form_mismatch` 派生:+- 选 a → `false`+- 选 b/c/d/e/f/g → `true`,cheat-status 持续提示"你的形态可能需要 bump 调权重"- **不再有"严重不匹配"档**——所有形态都能跑工作流,只是有的 rubric 需要更激进的 bump**Q1.5: 典型时长**(仅 Q1=a/d/f 时问)@@ -163,7 +177,13 @@ allowed-tools: Bash(*), Read, Write, Edit, Glob, WebFetch, Skill> 第一次 publish 之前装上就行(cheat-status 会持续提醒装)。> 装的指引在 adapters/perf-data/<platform>/README.md。"-记录到 `data_collection`(manual / adapter)。+**Q3 → `data_collection` enum 映射**:++| 用户答 | `data_collection` 写入值 |+|---|---|+| a | `"manual"` |+| b(默认) | `"adapter"` |+默认推荐 b——除非用户明确说 "a 我就要手动"。**Q4: 候选选题**@@ -173,7 +193,13 @@ allowed-tools: Bash(*), Read, Write, Edit, Glob, WebFetch, Skill> b) 有,markdown 列表> c) 有,Notion / 其他"-记录到 `pool_status`(none / markdown / notion)。+**Q4 → `pool_status` enum 映射**:++| 用户答 | `pool_status` 写入值 |+|---|---|+| a(默认) | `"none"` |+| b | `"markdown"` |+| c | `"notion"` |**Q5: 装几个 hook(默认装,不需要你决定)**@@ -191,7 +217,13 @@ allowed-tools: Bash(*), Read, Write, Edit, Glob, WebFetch, Skill>> 回 yes / no。"-记录到 `hooks_installed`(true / false)。+**Q5 → `hooks_installed` 映射**:++| 用户答 | `hooks_installed` 写入值 |+|---|---|+| yes / enter / 默认 | `true`(bool,**不是字符串 `"yes"`**) |+| no | `false` |+默认 yes——除非用户明确说 no。### Phase 2.5: 对标账号(**所有用户都问**,cold-start 强烈建议)@@ -234,32 +266,39 @@ c) 不找 → state 标 `benchmark_status: none`,用通用 v0 起步"正在创建 .cheat-state.json — 各子 skill 共享上下文的地方。这次 init 收集的所有答案都会写在这里。"```- 写入:+ 写入(**所有 `<...>` 占位必须查上面 Q 的映射表换成具体 enum 值,绝不直接存字母**):```json{"schema_version": "1.1","skill_version": "1.0.0","rubric_version": "v0",- "content_form": "<Q1>",- "typical_duration_seconds": <Q1.5 派生>,- "target_publish_cadence_days": <Q1.6 派生>,- "rubric_form_mismatch": <Q1≠a 时 true>,- "benchmark_status": "<none/imported/pending 派生自 Phase 2.5 答案>",- "benchmark_name": <如有 imported 则名字字符串,否则 null>,- "benchmark_sample_count": <如有 imported 则数字,否则 0>,+ "content_form": "<查 Q1 映射表,写 enum 字符串如 \"opinion-video\">",+ "typical_duration_seconds": <Q1.5 派生:30/90/240/450/900>,+ "target_publish_cadence_days": <Q1.6 派生:1/2/7/null>,+ "rubric_form_mismatch": <Q1=a→false;其他→true>,+ "benchmark_status": "<Phase 2.5 派生:a→\"imported\"/b→\"pending\"/c→\"none\">",+ "benchmark_name": <imported 则字符串名,否则 null>,+ "benchmark_sample_count": <imported 则数字,否则 0>,"baseline_plays": null,- "calibration_samples": <Q2=a 时 0;Q2=b 时 import 后回填或 Q2.3 估值>,- "data_collection": "<Q3>",- "pool_status": "<Q4>",+ "calibration_samples": <Q2=a→0;Q2=b→Phase 3.5 import 回填或 Q2.3 估值>,+ "data_collection": "<查 Q3 映射表,写 \"manual\" 或 \"adapter\">",+ "pool_status": "<查 Q4 映射表,写 \"none\"/\"markdown\"/\"notion\">","data_layer": "markdown",- "hooks_installed": <Q5>,+ "hooks_installed": <查 Q5 映射表,写 bool true/false>,"enabled_trend_sources": ["manual-paste"],- "enabled_perf_adapters": [],+ "enabled_perf_adapters": <Q2.1=a→[\"douyin-session\"];b→[\"youtube-data-api\"];c→[\"bilibili-stat\"];其他→[]>,+ "last_bump_at": null,+ "last_bump_self_audited": false,+ "last_published_at": null,+ "last_published_file": null,+ "last_retro_at": null,+ "last_trends_run_at": null,+ "last_trends_added_count": 0,"consecutive_directional_errors": [],"pending_retros": [],"shoots": [],"in_progress_session": null,- "initialized_at": "<ISO>"+ "initialized_at": "<本地 ISO 8601 含时区,如 \"2026-05-05T20:11:13+08:00\",**不要用 UTC 的 Z 后缀**>"}```@@ -425,7 +464,7 @@ cheat-learn-from 完成后回到 init 的 Phase 5。| `schema_version` | Phase 3 | 硬编码 "1.1" || `skill_version` | Phase 3 | 硬编码 "1.0.0" || `rubric_version` | Phase 3 | "v0" |-| `content_form` | Phase 3 | Q1 |+| `content_form` | Phase 3 | Q1 → 查映射表换 enum 值(**不是字母**) || `typical_duration_seconds` | Phase 3 | Q1.5 派生 || `target_publish_cadence_days` | Phase 3 | Q1.6 派生 || `rubric_form_mismatch` | Phase 3 | Q1≠a → true |@@ -434,7 +473,11 @@ cheat-learn-from 完成后回到 init 的 Phase 5。| `benchmark_sample_count` | Phase 3 / 2.5 | cheat-learn-from import 后回填 || `baseline_plays` | Phase 3.5(如 import 成功) | import 数据中位数;否则 null || `calibration_samples` | Phase 3 / Phase 3.5 | Q2=a→0;Q2=b→Q2.3 估值或 import 数 |-| `data_collection` | Phase 3 | Q3 |-| `pool_status` | Phase 3 | Q4 |-| `hooks_installed` | Phase 3-4 | Q5 + Phase 4 测试结果 |-| `initialized_at` | Phase 3 | now() |+| `data_collection` | Phase 3 | Q3 → 查映射表换 enum 值 |+| `pool_status` | Phase 3 | Q4 → 查映射表换 enum 值 |+| `enabled_perf_adapters` | Phase 3 | Q2.1 派生(如 Q2=a 则 `[]`) |+| `hooks_installed` | Phase 3-4 | Q5 → bool(不是字符串) |+| `last_bump_at` / `last_published_at` / `last_published_file` / `last_retro_at` / `last_trends_run_at` | Phase 3 | 全部 `null` |+| `last_bump_self_audited` | Phase 3 | `false` |+| `last_trends_added_count` | Phase 3 | `0` |+| `initialized_at` | Phase 3 | now() 本地 ISO 8601,含 `+08:00` 时区,**不要 UTC `Z`** |