Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →资本市场尽职调查 Skill：公司章程及组织机构（三会运作）核查（对应《编报规则第 12 号》第 11 章）。 当用户提到以下场景时触发：公司章程合规性、三会议事规则、股东大会议事规则、董事会议事规则、 监事会议事规则、独立董事制度、董事会专门委员会（审计、战略、提名、薪酬）、关联交易决策制度、 对外担保决策制度、对外投资决策制度、信息披露管理制度、内控制度、三会召开程序、表决程序、 会议通知、会议记录、决议有效性、公司治理结构、corporate governance、charter DD 等。 典型输入：现行公司章程及历次修订版、三会议事规则、独立董事工作制度、董事会专门委员会 实施细则、内控制度文件、近 3 年股东会 / 董事会 / 监事会的通知 / 议案 / 决议 / 会议记录 / 签到表。 非触发边界：设立时的章程归 ecm-dd-establishment；历次章程修订伴随的股本变动归 ecm-dd-history； 董监高任职资格归 ecm-dd-directors；股东身份核查归 ecm-dd-shareholders；本次发行的批准 决议归 ecm-dd-approv
.claude/skills/thomasmoreai-ecm-dd-charter/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 73% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 47% | 0% |
本 skill 负责：
本 skill 不负责：
ecm-dd-establishment）ecm-dd-history）ecm-dd-approval）ecm-dd-directors）ecm-dd-shareholders）见 DISCLAIMER.md。
执行本 skill 时，必须同时遵循 senior-lawyer-execution-standards.md。本 skill 的任何输出不得突破四条底线：事实可追溯、法源可核验、风险可分级、建议可落地；无法核验时必须显式标注。
ecm-setup-file-organize02-尽职调查/02-10-公司治理/charter差异；三会近 3 年召开是否符合章程和议事规则
| 级别 | 定义 | 本 skill 场景的典型例子 | |------|------|------------------------| | 高（红） | 实质性障碍 | 章程条款与《公司法》强制性规定冲突（如限制股东法定权利）、报告期内重大决议程序违法（如应经股东大会未经股东大会）、近 3 年全部未召开股东大会 | | 中（黄） | 瑕疵但可补正 | 议事规则部分条款缺失、独立董事未按期述职、董事会专门委员会未按期开会、内控制度文件未及时更新、会议记录要素不全 | | 低（蓝） | 形式瑕疵 | 章程个别表述与最新监管指引不完全一致、会议文件归档顺序混乱、签到表复印件盖章轻微模糊 |
严格遵循 shared/templates/dd-skill-template.md。 落地：02-尽职调查/02-10-公司治理/DD-Memo-公司治理-{YYYYMMDD}.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | fail→pass | 16,075 | 13,837 | -14% | 1 | 1 | 0% | 2,605 | 3,225 | +24% | 0 | 0 | — |
case-01 | fail→fail | 30,723 | 37,674 | +23% | 1 | 1 | 0% | 4,624 | 7,236 | +56% | 0 | 0 | — |
case-02 | fail→pass | 21,271 | 30,048 | +41% | 1 | 1 | 0% | 3,466 | 5,985 | +73% | 0 | 0 | — |
case-03 | fail→pass | 29,557 | 37,436 | +27% | 1 | 1 | 0% | 4,486 | 7,225 | +61% | 0 | 0 | — |
case-04 | fail→pass | 21,431 | 21,576 | +1% | 1 | 1 | 0% | 3,297 | 4,694 | +42% | 0 | 0 | — |
case-05 | fail→pass | 18,471 | 21,956 | +19% | 1 | 1 | 0% | 3,161 | 4,633 | +47% | 0 | 0 | — |
case-06 | fail→fail | 23,012 | 27,463 | +19% | 1 | 1 | 0% | 3,661 | 5,151 | +41% | 0 | 0 | — |
case-07 | fail→pass | 21,945 | 18,777 | -14% | 1 | 1 | 0% | 3,333 | 3,556 | +7% | 0 | 0 | — |
case-08 | fail→fail | 21,707 | 30,280 | +39% | 1 | 1 | 0% | 3,405 | 5,992 | +76% | 0 | 0 | — |
case-09 | pass→pass | 19,596 | 11,945 | -39% | 1 | 1 | 0% | 2,885 | 2,846 | -1% | 0 | 0 | — |
case-11 | pass→pass | 12,890 | 12,038 | -7% | 1 | 1 | 0% | 1,903 | 2,910 | +53% | 0 | 0 | — |
case-12 | fail→pass | 18,314 | 12,762 | -30% | 1 | 1 | 0% | 2,590 | 3,054 | +18% | 0 | 0 | — |
case-13 | pass→pass | 18,328 | 18,701 | +2% | 1 | 1 | 0% | 2,870 | 3,989 | +39% | 0 | 0 | — |
case-14 | pass→pass | 10,969 | 4,245 | -61% | 1 | 1 | 0% | 1,758 | 1,753 | -0% | 0 | 0 | — |
case-15 | pass→pass | 16,133 | 15,231 | -6% | 1 | 1 | 0% | 2,308 | 3,264 | +41% | 0 | 0 | — |
case-16 | pass→pass | 21,944 | 22,502 | +3% | 1 | 1 | 0% | 3,589 | 4,702 | +31% | 0 | 0 | — |
case-17 | pass→pass | 16,382 | 17,863 | +9% | 1 | 1 | 0% | 2,621 | 3,631 | +39% | 0 | 0 | — |
case-18 | fail→pass | 16,523 | 24,984 | +51% | 1 | 1 | 0% | 2,609 | 5,063 | +94% | 0 | 0 | — |
case-19 | fail→pass | 13,830 | 3,714 | -73% | 1 | 1 | 0% | 2,149 | 1,549 | -28% | 0 | 0 | — |
case-20 | pass→pass | 11,614 | 9,848 | -15% | 1 | 1 | 0% | 1,678 | 2,436 | +45% | 0 | 0 | — |
case-21 | pass→fail | 21,292 | 20,834 | -2% | 1 | 1 | 0% | 3,406 | 4,528 | +33% | 0 | 0 | — |
case-22 | pass→pass | 22,074 | 15,815 | -28% | 1 | 1 | 0% | 2,985 | 3,667 | +23% | 0 | 0 | — |
case-23 | fail→pass | 15,458 | 19,047 | +23% | 1 | 1 | 0% | 2,237 | 3,985 | +78% | 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. 23 cases were attempted. The headline lift of +39 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.