| name | finance-core-analysis |
| description | Generate stage-2 deep financial mechanism analysis markdown from a daily-finance brief plus current web verification. Use this skill whenever the user wants deeper macro/market analysis beyond daily news, asks for "核心分析"、"深度分析"、"机制分析"、"为什么市场这样走"、"帮我分析一下背后逻辑", or when the user references a daily-finance output and wants the second step of the pipeline before writing a 公众号爆款文章. Even if the user only says "深挖一下今天的市场" or "帮我分析这份财经简报", use this skill. This skill feeds finance-explosive-article as stage 2 of: daily-finance → finance-core-analysis → finance-explosive-article. |
Core Financial Analysis
Overview
Generate a deep mechanism analysis grounded in stage-1 daily finance facts plus live-verified data.
Pipeline position:
daily-finance → finance-core-analysis → finance-explosive-article
Your output is a standalone, publishable 公众号 deep-analysis article — rigorous enough to stand alone, structured enough to feed the final explosive article downstream.
Input
Preferred: Read markdown/daily-finance-YYYY-MM-DD.md
Fallback order:
- Ask for the report date or file path if ambiguous
- If user pastes raw daily-finance content in chat, use that
- Never invent missing facts
Step 1: Verify & Update Data
Always access current external data before writing.
Use web search / browser / MCP tools to:
- Re-check major market prices, yields, FX rates, commodities, volatility indices
- Verify policy statements, macro releases (CPI/PMI/jobs), earnings, geopolitical facts
- Update stale numbers from the stage-1 brief when newer reliable data is available
Source hierarchy:
- Primary: Official releases, central banks, exchanges, regulators, company filings; Reuters, Bloomberg, FT, WSJ
- Chinese primary: 财新, 第一财经, 财经杂志, 21世纪经济报道, 经济观察报
- Secondary (require source backtracking): FT中文网, WSJ中文网, 新浪财经
Only add data that is source-verifiable. Never fabricate.
Step 2: Apply the Core Model
Explain market behavior through these variables:
| Variable | What to examine |
|---|
| 流动性 Liquidity | Money supply, repo rates, bank reserves, credit conditions |
| 利率 Interest rates | Real vs nominal; short vs long end; central bank signals |
| 风险偏好 Risk appetite | VIX, credit spreads, EM flows, positioning |
| 资本流动 Capital flows | Cross-border flows, DXY, CNY pressure, EM vs DM |
| 政策方向 Policy | Fiscal, monetary, regulatory — direction and credibility |
| 资产负债表压力 Balance sheets | Corporate leverage, bank stress, sovereign debt |
| 激励约束 Incentive constraints | What is forcing or preventing capital allocation |
Causal chain template:
事件 → 约束变化 → 资金行为 → 定价结果 → 风险信号
Use this chain to anchor every mechanism explanation.
Step 3: Validate Before Writing
Check every item:
- Data: Dates correct? Units right? Directions accurate? Actual vs expected labeled? Intraday vs close distinguished?
- Sources: Important claims have reliable sources?
- Logic: Every conclusion follows from a mechanism, not mood words?
- Causal chain: Event → constraint → capital behavior → pricing result → risk signal present?
- Counter-case: Does the analysis include what would prove it wrong?
- Publishability: Title, subheadings, short paragraphs, clear conclusion?
Output Format
1. Title
公众号-style: tension + mechanism. Prefer substance over clickbait.
2. 核心判断(Executive Judgment)
One paragraph: the single most important market judgment today, and what specifically changed.
3. 事实基础(Fact Base)
- Inherit 3–5 key facts from daily-finance with source labels 【实】【预】【估】【待】
- Add newly verified data where necessary, clearly marked
- No invented facts
4. 机制分析(Mechanism Analysis)
Analyze through 3–5 relevant lenses from the core model. For each lens:
- What changed
- Why it matters
- How it transmits into asset prices
Use "不是A,而是B" contrast structure where helpful to clarify mechanism.
5. 核心矛盾(Core Contradiction)
Name the central tension driving the current market, e.g.:
- 增长 vs 通胀
- 政策宽松 vs 汇率压力
- 风险偏好修复 vs 盈利压力
- 流动性改善 vs 资产负债表收缩
6. 情景推演(Scenario Deduction)
| 情景 | 触发条件 | 资产含义 |
|---|
| 基准情景 | ... | ... |
| 备选情景 | ... | ... |
| 证伪信号 | ... | ... |
7. 关键跟踪变量(Variables to Watch)
3–6 observable indicators for the next update cycle.
8. 方向性参考(Non-Advisory Implication)
Explain directional exposure and risk. No ticker calls, no buy/sell recommendations.
9. 数据来源
Inherited sources from daily-finance + any new verified sources, one per line.
10. 免责声明
本文仅供参考,不构成投资建议。
File Output
Save to:
markdown/finance-core-analysis-YYYY-MM-DD.md
- Use the same report date as the input daily-finance file
- Create
markdown/ directory if missing
- UTF-8 encoding
- Output a complete publishable article, not notes
- Fallback: If write fails, output full markdown in chat
Writing Style
- Strong title, short paragraphs, numbered sections
- Sharp but rational tone — serious analysis, not hype
- Mechanisms explained in plain Chinese
- Use contrast: "不是A,而是B"
- No empty emotional phrases, no unexplained jargon
- Publishable as-is on 公众号
Goal
Answer four questions every reader should leave with:
- What changed — the key fact
- Why it changed — the mechanism
- How facts connect to asset pricing — the transmission
- What would prove this wrong — the falsification signal