Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Decode HOA covenants (CC&Rs) and the fee structure before you buy into them. Use when someone asks 'what do these HOA rules actually mean', 'decode these CC&Rs', 'is this HOA going to be a problem', or 'what should I check before buying in an HOA'. Produces a restriction decode ranked by lifestyle impact, special-assessment exposure analysis, enforcement and fine mechanics, and the exact records to request before buying.
.claude/skills/mohitagw15856-hoa-decoder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 83% | 0% |
> hoa-decoder 的简体中文翻译 — 英文版本为规范版本。
买进一个HOA,等于加入一个对你拥有征税权的微型政府。这个技能像一位在业委会会议上坐过很多次的朋友那样读CC&Rs:哪些规则会改变你的生活方式、意外账单能有多大、以及该先索要哪些记录。
仅在未提供时索取:
按以下主线通读文件:使用限制、摊派机制(月费上涨上限;批准门槛;未写上限 = 警示)、出租限制(配额、排队、最短租期、既有权利保留)、执行链条(通知 → 听证 → 罚款 → 留置;逐步引用原文)、修改规则(规则多容易在你脚下改变)。当条款的可执行性存疑时,标注"可执行性因司法辖区而异——请本地核实",而不是宣布其无效。
1. 结论——放心买 / 睁着眼买 / 这个HOA会和你的生活方式作对——三句话。
2. 按对你的影响排序的限制条款
| 限制(§) | 条款怎么写 | 如何冲击你的计划 | 严重程度 | |---|---|---|---|
3. 💸 金钱敞口——今天的月费、上涨机制、特别摊派规则引用、现实的最坏账单;罚款表和留置路径。
4. 🚩 警示信号,分级排序——引用的措辞、咬人的情景、严重程度。
5. 买前要索取的记录——储备金研究(及其充足率)、24个月业委会纪要、预算和欠费率、总保单、过去/未决的特别摊派与诉讼、出租配额现状和排队情况。
6. 问业委会/物业经理的问题——费用上涨历史("过去5年每年的月费是多少?")、即将到来的大修、罚款实际执行的频率。
在产出物结尾原文附上:"这是一份大白话解读,不构成法律/财务建议——法律因司法辖区而异;任何关键事项请向合格专业人士确认。"
买方侧HOA尽职调查实践——CC&R分诊、摊派敞口分析、记录清单。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 25,005 | 25,299 | +1% | 1 | 1 | 0% | 4,035 | 5,350 | +33% | 0 | 0 | — |
case-02 | fail→pass | 25,949 | 23,553 | -9% | 1 | 1 | 0% | 3,890 | 4,910 | +26% | 0 | 0 | — |
case-03 | fail→pass | 26,332 | 17,992 | -32% | 1 | 1 | 0% | 4,429 | 4,255 | -4% | 0 | 0 | — |
case-04 | pass→pass | 15,099 | 12,649 | -16% | 1 | 1 | 0% | 2,781 | 3,449 | +24% | 0 | 0 | — |
case-05 | pass→pass | 13,033 | 12,414 | -5% | 1 | 1 | 0% | 2,328 | 3,411 | +47% | 0 | 0 | — |
case-06 | pass→pass | 9,326 | 13,510 | +45% | 1 | 1 | 0% | 1,701 | 3,190 | +88% | 0 | 0 | — |
case-07 | pass→pass | 14,596 | 11,532 | -21% | 1 | 1 | 0% | 2,378 | 3,142 | +32% | 0 | 0 | — |
case-08 | pass→pass | 8,764 | 12,360 | +41% | 1 | 1 | 0% | 1,476 | 3,339 | +126% | 0 | 0 | — |
case-09 | pass→pass | 11,759 | 13,448 | +14% | 1 | 1 | 0% | 1,832 | 3,318 | +81% | 0 | 0 | — |
case-10 | pass→pass | 12,510 | 14,949 | +19% | 1 | 1 | 0% | 2,154 | 3,559 | +65% | 0 | 0 | — |
case-11 | pass→pass | 12,448 | 14,124 | +13% | 1 | 1 | 0% | 1,997 | 3,505 | +76% | 0 | 0 | — |
case-12 | fail→pass | 13,293 | 11,205 | -16% | 1 | 1 | 0% | 2,155 | 2,992 | +39% | 0 | 0 | — |
case-13 | fail→pass | 13,482 | 17,989 | +33% | 1 | 1 | 0% | 2,157 | 3,942 | +83% | 0 | 0 | — |
case-14 | fail→pass | 12,481 | 8,931 | -28% | 1 | 1 | 0% | 2,003 | 2,488 | +24% | 0 | 0 | — |
case-15 | pass→pass | 21,150 | 15,525 | -27% | 1 | 1 | 0% | 2,100 | 3,697 | +76% | 0 | 0 | — |
case-16 | fail→pass | 12,279 | 15,161 | +23% | 1 | 1 | 0% | 1,973 | 3,711 | +88% | 0 | 0 | — |
case-17 | fail→fail | 17,130 | 16,398 | -4% | 1 | 1 | 0% | 1,547 | 2,917 | +89% | 0 | 0 | — |
case-18 | pass→pass | 21,184 | 20,271 | -4% | 1 | 1 | 0% | 469 | 2,147 | +358% | 0 | 0 | — |
case-19 | pass→pass | 14,095 | 17,639 | +25% | 1 | 1 | 0% | 2,164 | 3,689 | +70% | 0 | 0 | — |
case-20 | fail→pass | 11,595 | 11,766 | +1% | 1 | 1 | 0% | 2,047 | 3,237 | +58% | 0 | 0 | — |
case-21 | pass→pass | 8,842 | 13,951 | +58% | 1 | 1 | 0% | 1,690 | 3,704 | +119% | 0 | 0 | — |
case-22 | pass→pass | 11,647 | 15,822 | +36% | 1 | 1 | 0% | 1,952 | 4,022 | +106% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +36 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.