shield_lock
金大哥
← 返回 Skill 列表
extension
分类: 数据与分析无需 API Key

每日热点信息分析报告

每日热点信息分析报告

person作者: user_d0c0f066hubcommunity
download下载资源包

每日热点信息分析报告（早盘版）

Overview

Generate a comprehensive daily hot topics analysis report covering 6 major Chinese social media platforms (Weibo, Zhihu, Baidu, Douyin, Toutiao, Bilibili), integrated with A-share market data analysis. The report is saved as a Markdown file in the current project directory.

When to Use

Use this skill when the user:

  • Explicitly asks to generate the daily hot topics report
  • Uses keywords: "每日热点"、"早盘报告"、"热搜分析"、"热点分析报告"、"hot report"
  • Asks for morning market sentiment analysis based on social media trends
  • This skill is triggered automatically by the 8:30 AM daily automation

Prerequisites

Before starting, determine:

  • Today's date (YYYYMMDD format) from the current system time
  • Last trading day: Try yesterday first, fall back to prior days (up to 7 days back)
  • Target file path: {当前项目目录}/每日热点信息分析报告_{YYYYMMDD}.md

STEP 1: Fetch Hot Search Data

Fetch data from all 6 platforms using WebFetch. If any platform fails, mark it as [获取失败] and continue.

| Platform | URL | TOP N | Heat Field | Notes | |----------|-----|-------|------------|-------| | Weibo (微博) | https://60s.viki.moe/v2/weibo | 20 | hot | | | Zhihu (知乎) | https://60s.viki.moe/v2/zhihu | 15 | hot | | | Baidu (百度) | https://60s.viki.moe/v2/baidu/hot | 20 | hot | | | Douyin (抖音) | https://60s.viki.moe/v2/douyin | 15 | hot | | | Toutiao (头条) | https://60s.viki.moe/v2/toutiao | 15 | hot | | | Bilibili (B站) | https://60s.viki.moe/v2/bili | 15 | None | Use rank weighting (see below) |

B站 scoring formula: B站 has no heat values. Use rank-based reverse weighting: Rank 1 = 75 pts, Rank 2 = 71 pts, Rank 3 = 67 pts... each rank decreases by 4 points. Formula: score = 75 - (rank - 1) * 4.

STEP 2: Fetch A-Share Market Data

Date strategy: At 8:30 AM, the market has not opened. Use the most recent trading day's data. Annotate in the report: "数据截至{YYYYMMDD}收盘". Find the nearest trading day by trying yesterday, then the day before, up to 7 days back.

Primary Path: finance-data API (Internal IDE API)

Use PowerShell with Invoke-RestMethod, POST requests, ContentType "application/json". API base URL: https://www.codebuddy.cn/v2/tool/financedata

2.1 Three Major Indices (MUST use index_daily, NOT daily)

# Shanghai Composite Index
$body = '{"api_name":"index_daily","params":{"ts_code":"000001.SH","start_date":"{YYYYMMDD_7days_ago}","end_date":"{YYYYMMDD_yesterday}"},"fields":"ts_code,trade_date,close,pct_chg,vol,amount"}'

# Shenzhen Component Index  
$body = '{"api_name":"index_daily","params":{"ts_code":"399001.SZ","start_date":"{YYYYMMDD_7days_ago}","end_date":"{YYYYMMDD_yesterday}"},"fields":"ts_code,trade_date,close,pct_chg,vol,amount"}'

# ChiNext Index
$body = '{"api_name":"index_daily","params":{"ts_code":"399006.SZ","start_date":"{YYYYMMDD_7days_ago}","end_date":"{YYYYMMDD_yesterday}"},"fields":"ts_code,trade_date,close,pct_chg,vol,amount"}'

⚠️ CRITICAL: Use index_daily for indices. The daily endpoint returns empty data for indices.

2.2 Limit-Up Sector Statistics (limit_cpt_list)

$body = '{"api_name":"limit_cpt_list","params":{"trade_date":"{last_trade_date}"},"fields":"trade_date,sector,up_cnt,limit_cnt,fd_amount,first_time,last_time,up_stat,limit_stat"}'

2.3 Limit-Up Stock List (limit_list_d)

$body = '{"api_name":"limit_list_d","params":{"trade_date":"{last_trade_date}","limit_type":"U"},"fields":"trade_date,ts_code,name,industry,close,pct_chg,limit_times,fc_ratio,fb_ratio,fd_amount,first_time"}'

2.4 Market Money Flow (moneyflow_mkt_dc)

$body = '{"api_name":"moneyflow_mkt_dc","params":{"trade_date":"{last_trade_date}"},"fields":"trade_date,buy_elg_amount,sell_elg_amount,buy_lg_amount,sell_lg_amount,buy_md_amount,sell_md_amount,buy_sm_amount,sell_sm_amount,net_amount"}'

⚠️ Unit is yuan (元). Convert to 亿元: divide by 1e8.

2.5 Northbound Capital Flow (moneyflow_hsgt)

# MUST pass both start_date and end_date to avoid getting stale historical data
$body = '{"api_name":"moneyflow_hsgt","params":{"start_date":"{5_days_before_last_trade}","end_date":"{last_trade_date}"},"fields":"trade_date,hgt,sgt,north_money,south_money"}'

⚠️ CRITICAL: north_money is cumulative position value (unit: 百万元). Calculate single-day net purchase as: 当日north_money - 前日north_money. ⚠️ This data has T+1 delay. Annotate in report: "前日数据（T+1）". ⚠️ Convert to 亿元: divide by 100.

2.6 Margin Trading (margin, optional)

$body = '{"api_name":"margin","params":{"trade_date":"{last_trade_date}"},"fields":"trade_date,exchange_id,rzye,rqye,rzmre,rzche,rqche"}'

⚠️ Unit is yuan (元). Convert to 亿元: divide by 1e8.

Fallback Path: East Money Public APIs

If ANY finance-data API call returns code:40204 or code:40101 (IP rate limit exceeded), IMMEDIATELY stop all finance-data calls and switch to East Money public APIs (all using WebFetch).

Fallback-A: Index Quotes + Capital Flow (single call)

URL: https://push2.eastmoney.com/api/qt/ulist.np/get?fltt=2&invt=2&fields=f1,f2,f3,f4,f12,f13,f14,f62,f184,f66,f69,f72,f75,f78,f81&secids=1.000001,0.399001,0.399006&ut=fa5fd1943c7b386f172d6893dbfba10b

Field mapping:

  • f2 = latest price, f3 = change %, f4 = change amount
  • f12 = code, f14 = name
  • f62 = main force net inflow (元), f66 = super-large order net inflow (元)
  • f72 = large order net inflow (元), f78 = medium order net inflow (元)
  • f184 = main force net inflow ratio (%)
  • Convert f62/f66/f72/f78 to 亿元: divide by 1e8

Fallback-B: Sector Rankings (East Money concept/industry sectors)

URL: https://push2.eastmoney.com/api/qt/clist/get?pn=1&pz=20&po=1&np=1&ut=bd1d9582d078c3fe4bd399c&fltt=2&invt=2&fid=f3&fs=m:90+t:2&fields=f2,f3,f4,f8,f12,f14,f62,f184&cb=&_={timestamp}

Take top 20 sectors by f3 (price change %). Fields: f14 = sector name, f3 = change %, f62 = main force net inflow (元).

Fallback-C: Northbound Capital (qualitative only before 9:30)

At 8:30 AM pre-market, northbound data is not available. Use qualitative assessment only. Annotate: "北向资金当日数据需9:30开盘后获取，以最近交易日定性判断代替。"

If finance-data API is available, use moneyflow_hsgt for T+1 historical data instead.

Fallback-D: Margin Trading (skip if finance-data is rate-limited)

Skip margin data when finance-data API rate limit is hit. Annotate in report: "[融资融券数据今日不可用（API超限），下一交易日补充]".

STEP 3: Generate Composite Hot Topics TOP20

Scoring formula per platform: Rank 1 = 100 pts, Rank 2 = 95 pts, each subsequent rank decreases by 5 pts.

B站 exception: Uses independent weighting — Rank 1 = 75, Rank 2 = 71, each rank decreases by 4 pts.

Merge logic: Match hot topics across platforms by keyword matching. Sum scores from all platforms a topic appears on. Sort by composite score descending, take TOP20.

STEP 4: Deep Analysis of TOP3-5 Hot Topics

For each of the top 3-5 hot topics in the composite ranking, produce a detailed analysis block. Each block MUST include all of:

  1. Event Summary (3-4 sentences): What happened, who is involved, what is the impact.
  2. Event Nature: Categorize as one of: 地缘政治 / 政策利好 / 政策利空 / 企业负面 / 科技突破 / 社会热点 / 行业事件 / 宏观经济 / 自然灾害 / 其他
  3. A-Share Impact Pathway: Classify as: 直接传导 / 情绪传导 / 预期传导 / 资金传导
  4. Related Sectors (list 2-4 with association rating ★):
    • ★★★★★ = Direct beneficiary / highly impacted
    • ★★★★ = Strong indirect impact
    • ★★★ = Moderate correlation
    • ★★ = Weak correlation
    • ★ = Peripheral
  5. Trading Suggestion: Include:
    • Allocation strength: 强 / 中 / 弱
    • Allocation horizon: 短期1-2天 / 中期1-2周 / 中长期1-3月
    • Key observation range: Critical price/valuation zone to monitor (no prediction)
    • Risk control notes: Key risk factors and contingency triggers

STEP 5: Generate the Report File

Save the report to: {当前项目目录}/每日热点信息分析报告_{YYYYMMDD}.md

Use the report template structure from assets/report_template.md as reference. The report has exactly 9 chapters:

Chapter Structure

  1. 各平台热搜综合排行 — Weighted TOP20 table with per-platform coverage indicators
  2. 重要热点事件深度解析 — TOP3-5 detailed analysis blocks (format from STEP 4)
  3. A股实盘行情 — Three major indices: close price, change %, volume (annotated with data date)
  4. 涨停板块分析 — Limit-up sector statistics + today's limit-up main theme judgment
  5. 资金面分析 — Market main force net inflow/outflow (亿元) + Northbound single-day net purchase (亿元, T+1) + Margin balance. If using fallback API, annotate data source.
  6. 热点→A股板块映射汇总表 — Mapping table: hot event → impact pathway → beneficiary/negative sectors → allocation suggestion
  7. 明日关注要点 — 3-5 items to watch tomorrow based on hot topic persistence and sector momentum
  8. 操作策略建议 — Position suggestion + focus directions + risk warnings
  9. 免责声明 — Fixed text: "本报告仅供参考，不构成投资建议。投资有风险，入市需谨慎。"

Critical Notes

  • B站 missing heat values: Normal behavior — use rank weighting as substitute. Note in report.
  • Empty API responses: Explicitly annotate [数据暂未入库，请见下一交易日] — do NOT silently skip.
  • Unit conversions: Market money flow/margin: divide by 1e8 (元→亿元); Northbound: divide by 100 (百万元→亿元).
  • Northbound cumulative value: north_money is cumulative. Single-day net = current day minus previous day.
  • Post-holiday: If today is the first trading day after a holiday, add 【节后首日】 tag to the report title.
  • Finance-data rate limit: When code:40204 or code:40101 is returned, MUST switch to East Money fallback APIs. Never leave the capital flow chapter empty.

Report Generation Workflow

  1. Execute STEP 1: Fetch all 6 platform hot search data in parallel
  2. Execute STEP 2: Fetch A-share market data (sequential API calls, but can use PowerShell in parallel where independent)
  3. Execute STEP 3: Compute composite scores and generate TOP20 ranking
  4. Execute STEP 4: Deep analysis of TOP3-5 topics
  5. Execute STEP 5: Assemble the full report using the template structure, write to file
  6. Present the generated file path to the user

Compliance and Style Rules

  • Prohibition: Never predict stock prices, index levels, or short-term movements. Never recommend specific stocks.
  • Data sourcing: Only use publicly available data. Annotate all data sources. Mark uncertainties explicitly.
  • Objectivity: Maintain neutral, objective tone. Distinguish facts from analysis.
  • Risk disclosure: Always include short-term, medium-term, and external sudden risks in Chapter 8.
  • Language: Professional, institutional-grade Chinese. No marketing language, no superlatives.
  • Disclaimer: Chapter 9 is mandatory and must use the fixed disclaimer text.