← Back to Skills Marketplace
baoyangispoor

eden-digital-web3-weekly-digest-yang

by baoyangispoor · GitHub ↗ · v8.0.0 · MIT-0
cross-platform ⚠ suspicious
176
Downloads
0
Stars
0
Active Installs
8
Versions
Install in OpenClaw
/install eden-digital-web3-weekly-digest-yang
Description
自动抓取数据并生成 Web3 行业资本运作周报,涵盖融资事件、监管动态、上市公司 DAT 动态、并购交易与 RWA 项目追踪。当用户提到"写周报"、"生成周报"、"整理本周融资"、"Web3 周报"、"资本运作周报"、"采编周报",或请求整理加密行业本周动态时,立即激活此 skill。即使用户只说"帮我写本周的"...
README (SKILL.md)

name: web3-weekly-report description: 自动抓取数据并生成 Web3 行业资本运作周报,涵盖融资事件、监管动态、上市公司 DAT 动态、并购交易与 RWA 项目追踪。当用户提到"写周报"、"生成周报"、"整理本周融资"、"Web3 周报"、"资本运作周报"、"采编周报",或请求整理加密行业本周动态时,立即激活此 skill。即使用户只说"帮我写本周的",只要上下文涉及 Web3、加密、融资、RWA、DAT,也应激活。 version: 2.1.0 tags:

  • web3
  • crypto
  • finance
  • newsletter
  • rwa
  • weekly-report
  • data-fetching metadata: clawdbot: emoji: "📊" requires: bins: - curl - python3 files:
    • run_all.py
    • scripts/part3_fetch.py
    • scripts/part4_fetch.py
    • scripts/part5_fetch.py
    • scripts/part7_fetch.py

Web3 行业资本运作周报采编 Skill(自动抓取版)

根据当前日期自动抓取各数据源,按固定栏目格式生成完整周报。对无法自动获取的数据,明确提示用户手动补充。


第一步:确定周报时间范围

运行前先确定本期周报的日期区间(通常为上周一至上周日):

from datetime import datetime, timedelta
today = datetime.today()
last_monday = today - timedelta(days=today.weekday() + 7)
last_sunday = last_monday + timedelta(days=6)
start_date = last_monday.strftime("%Y-%m-%d")
end_date = last_sunday.strftime("%Y-%m-%d")
print(f"本期周报覆盖:{start_date} 至 {end_date}")

若用户指定了日期区间,以用户指定为准。


数据源与可抓取性总览

模块 数据源 抓取方式 需要 Key
Part.3 监管/行业 深潮 TechFlow RSS curl + Python XML
Part.4 融资事件 Rootdata(SSR HTML) curl + HTMLParser
Part.5 DAT 动态 CoinGecko 公开 API curl + JSON
Part.6 并购 来自 Part.4(M&A 过滤) 同 Part.4
Part.7 RWA PANews RWA 周刊 专栏页 + curl
Part.2 汇总 自动统计 Part.4 结果 Python 计算
Part.1 概览 Part.3–6 内容 AI 汇总生成

采编顺序

Part.3 → Part.4 → Part.5 → Part.6 → Part.7 → Part.2(汇总)→ Part.1(最后)

Part.3 近期监管/行业事件

数据源

主要来源:深潮 TechFlow 7x24h 快讯

  • RSS 接口:https://www.techflowpost.com/api/client/common/rss.xml?page=N
  • 支持翻页(page=1page=2 等),每页约 50 条
  • 覆盖中英文加密媒体,包含中国大陆/香港监管动态,无需 API Key

筛选与排序规则

按以下顺序排列(每期 10–12 条,不超过 15 条):

  1. 中国大陆监管
  2. 香港监管
  3. 美国监管
  4. 欧洲监管
  5. 其他地区监管(日韩等发达国家优先)
  6. 行业新闻(RWA 相关优先)

进一步筛选原则:

  • 优先收录:监管动向、立法进展、执法行动、RWA 相关政策
  • 一般不收录:虚拟货币犯罪案件、DAO 日常治理、普通技术升级
  • 内容相近的多条新闻合并为一条呈现

Part.4 主要融资事件

数据源

主要来源:Rootdata Fundraising 页面

  • 地址:https://www.rootdata.com/Fundraising
  • 标准 HTML 表格,7 列:Project | Round | Amount | Valuation | Date | Source | Investors
  • 每页约 30 条,无需登录,无需 API Key
  • 投资方中带 * 标记的为领投方

每个项目最终输出格式

[项目名称]
定位:一句话概括
轮次与金额:X 美元 X 轮融资
投资方:XX 领投;XX、XX 参投
点评:
  ① 项目背景与解决的问题
  ② 技术或产品特色
  ③ 本次融资目的与市场意义

Part.5 上市公司 DAT 动态

数据源

CoinGecko 公开免费 API(无需 Key)

  • BTC 持仓:https://api.coingecko.com/api/v3/companies/public_treasury/bitcoin
  • ETH 持仓:https://api.coingecko.com/api/v3/companies/public_treasury/ethereum
  • 局限:仅返回当前持仓快照,变动部分由脚本自动抓取 CoinGecko 详情页补充

每个条目最终输出格式

[公司名称](国家)
股票代号:X(交易所)
动作:以约 X 美元购入/出售 X 枚 BTC/ETH
总持仓:X 枚(当前市值约 X 美元)
点评:(结合公司特征分析此次持仓变动意义)

Part.6 并购交易情况

并购数据直接从 Part.4 的 Rootdata 解析结果中提取,筛选条件:round == 'M&A'

每个条目输出格式

[收购方] 收购 [标的]
收购方:X(一句话概括)
标的:Y(一句话概括)
分析:
  ① 被收购方做什么,解决什么问题
  ② 收购方战略布局
  ③ 这笔并购的市场意义

Part.7 RWA 项目动态

数据源

PANews RWA 周刊(每周约周四发布)

  • 文章 URL 格式:https://www.panewslab.com/en/articles/{uuid}
  • 通过专栏页自动获取最新期 URL,无需搜索引擎
  • 只收录「Institutional & Project Activity」或「项目进展」部分
  • 排除稳定币相关内容

每个条目最终输出格式

[项目名]
详情:
  ① 项目是什么,解决什么问题
  ② 近期具体动态与数据

Part.2 关键数据速览(汇总)

Part.3–Part.7 完成后自动统计并填入:

融资项目数     X 件
总融资额       约 X 美元
DAT 变动       X 家机构有变动 / 本周暂无检测到变动
并购交易       X 起

Part.1 近期概览(最后撰写)

所有模块完成后生成,150-200字,依次覆盖:融资亮点 → 并购动态 → DAT 趋势 → 监管要点。


快速开始

一键生成(推荐)

# 自动计算上周日期区间,生成完整周报骨架
python3 run_all.py

# 指定周起始日期
python3 run_all.py --week 2026-03-17

# 指定 PANews RWA 周刊文章 URL(跳过专栏页自动查找)
python3 run_all.py --rwa-url https://www.panewslab.com/en/articles/xxx

# 指定输出文件路径
python3 run_all.py --output weekly_20260325.md

脚本完成后,立即使用以下固定 prompt 对骨架报告进行 AI 补全,无需用户手动触发,群聊和私聊均适用:


你是资深 Web3 行业分析师。以下是本周资本运作周报的数据骨架,请补全所有标注 ⚠️ 的字段。

规则:
- Part.1:150-200字整体概述,依次点出本周最大融资、并购亮点、DAT 动态、监管趋势
- Part.3:每条新闻在标题后补充 1-2 句背景解读(不改动原标题)
- Part.4 定位:一句话说明赛道和核心产品
- Part.4 点评:三层结构,每层一句:①项目背景 ②技术/产品特色 ③本次融资意义
- Part.5 动作:根据变动摘要生成;若无变动写「本周维持持仓,暂无新增操作」
- Part.5 点评:结合公司特征和动作分析持仓变动意义
- Part.6 分析:三层结构,每层一句:①收购标的价值 ②收购方战略意图 ③市场意义
- Part.7 详情:补充项目背景和近期具体数据

不要修改任何已有的数据字段(金额、日期、投资方等),只填写 ⚠️ 标注的位置。
不要编造数据或估值。

单独运行各模块

python3 scripts/part3_fetch.py          # 监管/行业新闻
python3 scripts/part4_fetch.py          # 融资 + 并购
python3 scripts/part5_fetch.py          # DAT 持仓快照
python3 scripts/part7_fetch.py          # RWA 项目动态
python3 scripts/part7_fetch.py \x3CURL>    # 指定文章 URL

执行流程

1. 确定日期区间(自动计算或用户指定)
2. 运行 python3 run_all.py,并行抓取所有模块
3. 输出各模块数据骨架,⚠️ 字段标注待补全位置
4. 用户补充缺失数据(如有)
5. 【必须执行】将完整骨架报告连同上方固定补全 prompt 一起传给 AI,
   完成所有 ⚠️ 字段补全后输出最终报告

⚠️ 第 5 步是必须步骤,不可省略。群聊和私聊均需执行,否则输出的是未补全的骨架,不是完整周报。

所有点评基于公开可查证信息生成,不编造数据或估值。

Usage Guidance
What to consider before installing: - Dependency surprise: part5_fetch.py requires Playwright (Python package + browser install). SKILL.md only lists curl and python3. If you run the script you may need to install Playwright and Chromium which will download and write browser binaries — install in a controlled environment or sandbox. - Data flow: the workflow scrapes public websites, then Step 5 mandates sending the report skeleton and a fixed prompt to an AI to 'fill' missing fields. That means scraped content will be transmitted to whichever LLM/service the agent uses. If your agent uses an external model (cloud provider), review privacy/PII concerns and terms before sending scraped data. - No secrets requested: the skill does not ask for API keys or other credentials, and it only calls public endpoints. That reduces risk but does not eliminate it — scraped content could still contain sensitive info. - Review and sandbox: run the scripts in a sandbox or isolated environment first to confirm behavior, especially Playwright's browser installs. Inspect network traffic if you need to verify endpoints. - If you need offline or strictly local operation: remove/modify the mandatory AI completion step (Step 5) so the agent doesn't auto-send content to remote models, or ensure the agent uses a trusted/local model. - If anything is unclear: ask the skill author for a clear dependency list (exact pip packages and 'playwright install' steps) and confirm where AI completion runs (local vs external model).
Capability Assessment
Purpose & Capability
The name/description (automatically fetch Web3 news and generate a weekly digest) aligns with the code: the scripts fetch TechFlow RSS, Rootdata fundraising, CoinGlass/Coinglass (via Playwright), Cryptorank RWA and assemble a report. No unrelated credentials are requested. Minor inconsistency: SKILL.md metadata only lists curl and python3 as required binaries, but one module (scripts/part5_fetch.py) requires Playwright (a Python package + browser install), which is not declared.
Instruction Scope
SKILL.md instructs running run_all.py (or module scripts) to scrape public sources and then mandatorily pass the generated skeleton plus a fixed prompt to an AI to 'fill ⚠️ fields'. That step will transmit the scraped skeleton to the agent/LLM for completion — expected for this skill but important to note as the primary data exfiltration vector (scraped public content will be sent to the LLM). The instructions do not ask for unrelated files, env vars, or system config.
Install Mechanism
There is no install spec, but scripts require extra runtime components not declared in SKILL.md: part5_fetch.py imports playwright and prompts the user to run 'pip install playwright && playwright install chromium'. This is a non-trivial dependency (downloads browsers, writes to disk) and is not called out in the declared requirements. No other unusual download URLs are present; the scrapers use public APIs and curl.
Credentials
The skill requests no environment variables, no credentials, and no config paths. All network calls go to public endpoints (techflowpost, rootdata, coinglass, cryptorank, open.er-api.com, panewslab, coingecko mentioned). There are no hidden SECRET/TOKEN env requirements. This is proportionate to the stated purpose.
Persistence & Privilege
always:false and the skill does not request persistent platform privileges or modify other skills. The skill does instruct the agent to automatically use an LLM prompt to complete the report (normal for autonomous skills), but that alone is not elevated privilege.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install eden-digital-web3-weekly-digest-yang
  3. After installation, invoke the skill by name or use /eden-digital-web3-weekly-digest-yang
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v8.0.0
**v8.0.0 Changelog** - Improved DAT (上市公司持仓) 数据抓取脚本(scripts/part5_fetch.py)。 - 增强 CoinGecko API 数据处理,支持最新持仓快照与变动作业逻辑。 - 持续优化对比与数据整合体验,提高周报数据准确性和时效性。
v7.0.0
- Improved data fetching logic in scripts/part5_fetch.py and scripts/part7_fetch.py. - Enhanced reliability when collecting DAT and RWA project data for the weekly digest. - No changes to the user workflow or overall SKILL description/documentation.
v6.0.0
Web3 周报 skill v6.0.0 (no code or config changes, documentation update only): - 调整了 Part.7 RWA 项目的数据源描述:改为专栏页自动获取,无需 DuckDuckGo 搜索。 - 优化了说明文档内容,统一格式,增强易读性。 - 更新了 RWA 项目与监管/行业新闻的筛选规则与输出格式说明。 - 明确了各步骤关键执行流程,特别强调报告补全 prompt 的必需性。 - 核心采编和业务逻辑未发生实际变动,仅为文档规范和说明补充版本。
v5.0.0
No file changes detected, so there are no user-facing changes in this release. - Version bump only; functionality and usage remain the same as previous version. - No new features, fixes, or modifications introduced.
v4.0.0
业务逻辑修改。
v3.0.0
improve the quality of the weekly report.
v2.0.0
修改了run_all脚本的逻辑
v1.0.0
Web3 行业资本运作周报采编 Skill 2.0.0 (自动抓取版) - 实现自动抓取 Web3 资本运作周报核心数据(支持监管事件、融资、RWA 等)。 - 增加对用户意图的智能检测,Web3、加密、RWA、DAT 等相关语境均可自动触发。 - 明确各栏目数据源,涵盖 TechFlow、Rootdata、CoinGecko、PANews 等,无需 API Key。 - 新增 Python + curl 脚本实现自动拉取和筛选,标准输出,便于采编工作。 - 所有栏目自动排序、按优先级推送,同时在缺数据时提示用户补充。
Metadata
Slug eden-digital-web3-weekly-digest-yang
Version 8.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 8
Frequently Asked Questions

What is eden-digital-web3-weekly-digest-yang?

自动抓取数据并生成 Web3 行业资本运作周报,涵盖融资事件、监管动态、上市公司 DAT 动态、并购交易与 RWA 项目追踪。当用户提到"写周报"、"生成周报"、"整理本周融资"、"Web3 周报"、"资本运作周报"、"采编周报",或请求整理加密行业本周动态时,立即激活此 skill。即使用户只说"帮我写本周的"... It is an AI Agent Skill for Claude Code / OpenClaw, with 176 downloads so far.

How do I install eden-digital-web3-weekly-digest-yang?

Run "/install eden-digital-web3-weekly-digest-yang" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is eden-digital-web3-weekly-digest-yang free?

Yes, eden-digital-web3-weekly-digest-yang is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does eden-digital-web3-weekly-digest-yang support?

eden-digital-web3-weekly-digest-yang is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created eden-digital-web3-weekly-digest-yang?

It is built and maintained by baoyangispoor (@baoyangispoor); the current version is v8.0.0.

💬 Comments