market-diagnostics SKILL.md skillA

market-diagnostics from wonglaitung/fortune is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add wonglaitung/fortune/market-diagnostics@git:20261101.c5a5005

Pins git:20261101.c5a5005 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
wonglaitung/fortune
Path
.opencode/skills/market-diagnostics/SKILL.md
Versions recorded
2
Latest version
git:20261101.c5a5005, committed Nov 1, 2026
First version recorded
Oct 30, 2026
Size
7,249 bytes
Repository stars
92
License
none declared

MUST USE when 需要回答“这个跑赢/跑输是真的吗、来自哪里”,或做指数 vs 个股分解、 IC/流动性/Beta/行业归因、成本校准、事件研究、幸存者检查、端点/权重敏感性。 触发词:归因/诊断/为什么跑赢/为什么亏/指数 vs 个股/幸存者/含息/成本/换手/事件研究/ 这个结论可信吗。本技能给出项目通用的**市场诊断 SOP + 硬规则 + 工具命令**; 不预测收益,只把收益/指标拆到可复核的口径。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# 市场诊断与归因 SOP(本项目的“诊断尺子”)

> 目标:把「看着跑赢」拆成 **股息 / 端点 / 权重 / 幸存者 / 成本 / 真实选股** 六件事。
> 一句话:**先问口径,再看数字;先拆来源,再谈能力。**

## 何时用

- 指数 vs 个股/组合:为什么十年涨/跌、谁贡献的;
- 某策略「超额为正」:是 alpha、beta、风格,还是成本/基准口径;
- IC 很高但不赚钱:信息落在哪个分层、能不能交易;
- 事件后漂移:盈警/配售/供股/卖空;
- 任何要作为「结论」输出的收益/超额数字。

不适用:预测模型训练(走 `hypothesis-validation`);个股基本面(走 `equity-analysis`)。

---

## 0. 先定口径(不做这步,后面全错)

| 问题 | 必须写死 |
|---|---|
| 基准 | 价格指数还是**含息**?等权还是市值加权? |
| 宇宙 | 当前池还是 PIT 池?含不含退市? |
| 收益 | 前复权?含股息?币种? |
| 端点 | 起止日;是否报多端点? |
| 成本 | 往返多少?flat 还是分层模型? |
| 显著性 | 行级还是折/月聚类? |

**数据缺口(本项目现状)**:
- 股数:FF301 月度只覆盖 30 只(2015–2026);310 只无完整官方历史;
- 退市:有除牌代码名单(810 条),**无逐日行情/退市日**;
- 指数:HSI PIT 成分 2023-07 起;无历史权重;
- 南向:逐股持仓可用(2024-10 起),更早只有总量。
⇒ 缺这些时,只能做**描述性诊断**,不能声称「无偏回测」。

---

## 1. 指数 vs 个股:六步拆解

1. **股息**:价格指数 vs 含息基准(如 2800.HK);十年股息差常达 30–40pp;
2. **端点**:至少报「全期 + 两个高点起算」;同一段历史换起点可翻符号;
3. **权重**:等权 vs 市值加权;少数巨头可主导指数;
4. **成分/风格**:老经济 vs 新经济、大小盘、价值/成长;
5. **幸存者**:当前池回填历史 ≠ 当时池;中位数比均值更诚实;
6. **成本/换手**:扣往返成本后再比较。

> 示例结论模板:「恒指价格 +5.0%,含息 +43.8%;310 当前池中位 +2.1%;
> 自选 30 +204% 但属事后挑选的赢家子集」——每句都要能复算。

## 2.5 筹码/股本(CCASS + FF301)

```bash
python3 scripts/ccass_concentration.py --date latest --top 20
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add wonglaitung/fortune/market-diagnostics@sha256:deea46d06a91bc03

The line at the top of this page pins the label git:20261101.c5a5005, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_gyfjlekiv72qxuwu.svg)](https://markdownregistry.com/a/art_gyfjlekiv72qxuwu)

0 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 2 versions of this file, between Oct 30, 2026 and Nov 1, 2026, about one every 2 days.

versioncommittedcommitsizeaudit
git:20261101.c5a5005 latest2026-11-01 c5a5005 7,249 BA view · diff
git:20261030.90a35192026-10-30 90a3519 6,126 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (7249 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

wonglaitung/fortune · 92 stars · license none declared · pushed 2026-11-03 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_gyfjlekiv72qxuwu
GET https://markdownregistry.com/api/v1/resolve?ref=wonglaitung/fortune/market-diagnostics
GET https://markdownregistry.com/api/v1/blob/deea46d06a91bc0318777753c8b7675a9434d692e9952cc834f1594f592f7be5

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from wonglaitung/fortune

equity-analysis SKILL.md skill
wonglaitung/fortune · .opencode/skills/equity-analysis/SKILL.md · MUST USE when 分析一只/一组股票的中长期投资价值,或回答“贵不贵 / 值不值得长期买 / 财务质量如何 / 怎么估值 / 定投还是分批”。触发词:分析股票/基本面/财报/估值/PE/PB/ROE/…
git:20261101.c5a5005 · audit A · 92 stars
hypothesis-validation SKILL.md skill
wonglaitung/fortune · .opencode/skills/hypothesis-validation/SKILL.md · MUST USE when 需要验证一个市场/策略假设是否成立,或判断某个指标/结论是否可信。 触发词:验证/证明/假设检验/回测/显著性/这策略行不行/这个能赚钱吗/有没有 alpha/…
git:20261101.c5a5005 · audit A · 92 stars
AGENTS.md AGENTS.md file
wonglaitung/fortune · AGENTS.md
git:20261101.2243443 · audit A · 92 stars

Every file in wonglaitung/fortune

Browse by kind, by grade A, or by owner.