← Back to Skills Marketplace
kenneth-bro

investoday-sector-research-interpretation

by investoday · GitHub ↗ · v1.3.0 · MIT-0
cross-platform ✓ Security Clean
173
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install investoday-sector-research-interpretation
Description
面向行业、板块、概念与主题方向的研报解读,聚焦卖方共识、核心逻辑、机会风险与代表性方向。基于今日投资金融数据接口,自动识别行业或概念实体并输出结构化板块研报解读报告。触发词:板块研报、行业观点、赛道共识、主题分歧、重点关注方向。
README (SKILL.md)

🧭 板块研报解读

面向行业、板块、概念与主题方向的研报解读,聚焦卖方共识、核心逻辑、机会风险与代表性方向。基于今日投资金融数据接口,自动识别行业或概念实体并输出结构化板块研报解读报告。

触发场景

  • 用户询问某个行业、板块、概念或主题最近机构怎么看
  • 用户希望了解卖方共识、核心逻辑、主要分歧和机会风险
  • 用户想知道“最近券商重点覆盖哪些方向”“这个板块分歧大不大”
  • 关键词:板块研报、行业观点、赛道共识、主题分歧、重点关注方向、板块研究

输入示例

示例 1:行业观点

半导体板块最近券商怎么看?

示例 2:共识与分歧

机器人板块最近研报里的共识和分歧是什么?

示例 3:机会风险

创新药这条赛道最近机构最看好的方向和风险点有哪些?

💡 本 Skill 偏行业、板块、概念层面的卖方观点归纳。若用户只想看单只股票的研报观点,请优先使用 个股研报解读;若用户更关心最近发生了什么催化、热点如何演绎,请优先使用 热点事件解码

前置依赖

本 Skill 依赖 investoday-finance-data(今日投资金融数据)Skill 获取实时金融数据。

基础 API 调用与底层执行方式统一以该 Skill 为准,业务 Skill 不重复展开底层接入细节。

工具说明

以下为本 Skill 通过 investoday-finance-data 使用的数据接口。在 System Prompt 中以 工具ID 标识调用。

对象识别工具

工具名称 工具ID 方法 说明
实体识别 entity-recognition POST 识别用户输入中的行业、概念、主题对象

研报核心工具

工具名称 工具ID 方法 说明
研报舆情 research/sentiment POST 获取行业、概念或主题方向的研报观点、逻辑、机会与风险

数据获取流程

用户提供行业、板块、概念或主题方向后,Agent 按以下流程获取数据:

  • Step 0:实体识别:工具ID entity-recognition (POST),参数 input=\x3C用户原始问题>
  • Step 1:行业/概念研报舆情:工具ID research/sentiment (POST),参数 \x3CindustryCode或conceptCode参数> beginTime=\x3C90天前> endTime=\x3C当前时间> pageNum=1 pageSize=6

对象选择规则:若识别结果为行业,优先传 industryCode;若识别结果为概念,传 conceptCode。默认使用近 90 天数据;只有用户明确要求“详细报告”时,pageSize 可放大到 10。

分析框架(4步)

Agent 获取数据后,按以下 4 步框架进行结构化分析:

Step 1:确认当前卖方总体态度

目标:先判断市场机构对该板块或主题的整体态度偏积极、偏中性还是分歧较大。

数据来源research/sentiment

分析要点:

  • 近期研报情绪方向是否集中
  • 研报覆盖是否充分
  • 若样本有限,应提示“结论代表性有限”

输出:总体态度与覆盖情况。

Step 2:提炼研报主线与共识逻辑

目标:归纳机构高频强调的主线逻辑。

数据来源research/sentiment

分析要点:

  • 当前最重要的产业逻辑、政策逻辑、景气逻辑或估值逻辑
  • 哪些逻辑是多数机构共同提及的
  • 只保留最有代表性的 2-4 条共识主线

输出:研报主线与卖方共识。

Step 3:识别分歧点与重点方向

目标:找出机构之间的主要分歧,以及反复被强调的重点方向。

数据来源research/sentiment

分析要点:

  • 分歧集中在兑现节奏、景气持续性、盈利能力还是估值容忍度
  • 哪些子方向或代表性公司被多次提及
  • 只有被多篇研报重复强化的方向,才可写成“重点关注方向”

输出:分歧点与重点方向。

Step 4:形成机会风险与后续观察框架

目标:把共识、分歧与重点方向整合为结构化观察框架。

数据来源:前 3 步分析结果汇总

分析要点:

  • 当前最值得关注的机会方向是什么
  • 当前最需要警惕的风险点是什么
  • 接下来应重点跟踪哪些验证信号

输出:机会与风险判断、后续观察重点。

策略逻辑汇总

信号组合 含义 判断
多篇研报逻辑同向且近 90 天持续强化 板块卖方共识较强 ✅ 积极
共识主线明确但子方向分歧较大 主线成立、细分选择有争议 🟡 关注
机会与风险同时被高频提及 机构表达趋于平衡 📊 中性
重点方向被多篇研报重复强化 方向识别度较高 ✅ 积极
研报样本少且观点分散 共识尚不稳定 ⚠️ 警惕
逻辑更多来自政策催化而非业绩兑现 更偏预期交易 🟡 关注
景气持续性成为主要分歧 后续验证难度较高 ⚠️ 警惕
风险点集中于估值与兑现节奏 需重点跟踪业绩验证 ⚠️ 警惕

输出格式

# 🧭 [板块/主题名称] 研报解读报告

> 分析日期:YYYY-MM-DD | 数据来源:今日投资

## 一、研报结论

(用一段话概括当前机构态度、核心主线、主要分歧和机会风险)

## 二、研报主线

(提炼 2-4 条高频主线逻辑)

## 三、共识与分歧

(说明共识逻辑与争议点)

## 四、重点方向

(说明最被看好的子方向或反复提及的代表方向)

## 五、机会与风险

(分别写机会方向和风险方向)

## 六、后续观察

(接下来需重点跟踪的验证信号)

## 综合结论

- 3-5 条核心发现
- 明确当前卖方共识强弱与关键分歧
- 给出后续研究的关键抓手

证据约束(必须遵守)

  1. 每个板块判断至少给出 2 个证据来源;没有数据则写“该维度数据不足,暂无法判断”
  2. 不允许把个别机构的单篇观点直接写成板块共识,必须有重复验证
  3. 时间口径必须明确,如“近90天研报”
  4. 不展示内部打分、未公开排序或中间推理
  5. 不给买卖建议、目标价、仓位建议或交易时点
  6. 只给行业、板块、概念、子方向层面的判断,不做具体个股推荐
  7. 若对象识别不稳定,必须先要求用户提供更明确的行业、板块或概念名称

执行示例

用户说:“机器人板块最近机构最看好什么方向?”

  1. 通过 entity-recognition 识别行业或概念对象
  2. 调用 research/sentiment 获取近 90 天相关研报舆情
  3. 提炼研报主线、共识与分歧、重点方向和风险点
  4. 输出 Markdown 格式板块研报解读报告
  5. 在结尾写出综合结论与后续观察重点

安全与隐私

  • 仅通过今日投资 API 查询公开市场数据
  • 不记录、不存储用户的查询记录
  • 分析结论仅供参考,不构成投资建议
Usage Guidance
This skill appears internally consistent and low-risk by itself. Before installing or enabling it for production use, review the investoday-finance-data skill it depends on (that skill will handle API keys and network access). Confirm that the underlying data skill's install, env vars, and storage/privacy behavior match your security requirements, and test with non-sensitive queries to verify the promise '不记录、不存储用户的查询记录' is honored in practice. If you need auditability, request logs or telemetry controls from the deployer of the investoday-finance-data tool.
Capability Analysis
Type: OpenClaw Skill Name: investoday-sector-research-interpretation Version: 1.3.0 The skill bundle is a financial analysis tool designed to interpret sector-level research reports using the 'investoday-finance-data' dependency. It follows a structured four-step analysis framework (attitude, consensus, divergence, and risk) and enforces strict constraints against providing specific investment advice or individual stock recommendations. There are no signs of data exfiltration, malicious execution, or harmful prompt injection; the logic is entirely focused on processing financial data into structured reports as described in SKILL.md.
Capability Assessment
Purpose & Capability
The skill claims to produce structured sector/theme research summaries and explicitly delegates data access to the investoday-finance-data skill. All required operations (entity recognition, research/sentiment calls) are declared and match the stated purpose; there are no unrelated environment variables, binaries, or config paths requested.
Instruction Scope
SKILL.md provides a clear 4-step analysis flow and specifies exact toolIDs and parameters (entity-recognition; research/sentiment with 90-day window). It does not instruct the agent to read local files, system state, or arbitrary environment variables, nor to transmit data to endpoints outside the declared tool. It also includes evidence and output constraints (e.g., at least 2 evidence sources, no stock-level recommendations).
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so nothing is written to disk or downloaded by the skill itself. That minimizes install-time risk.
Credentials
The skill declares no required environment variables, credentials, or config paths. All API access is routed through the investoday-finance-data skill; any credential needs would arise from that dependent skill, not this one.
Persistence & Privilege
always:false and user-invocable:true. The skill does not request permanent presence or modification of other skills or system-wide settings. There is no instruction to persist or cache user queries beyond a declarative privacy note in the doc.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install investoday-sector-research-interpretation
  3. After installation, invoke the skill by name or use /investoday-sector-research-interpretation
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.3.0
Automated release from GitHub Actions
v1.1.0
Manual release
v1.0.0
Manual release
Metadata
Slug investoday-sector-research-interpretation
Version 1.3.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is investoday-sector-research-interpretation?

面向行业、板块、概念与主题方向的研报解读,聚焦卖方共识、核心逻辑、机会风险与代表性方向。基于今日投资金融数据接口,自动识别行业或概念实体并输出结构化板块研报解读报告。触发词:板块研报、行业观点、赛道共识、主题分歧、重点关注方向。 It is an AI Agent Skill for Claude Code / OpenClaw, with 173 downloads so far.

How do I install investoday-sector-research-interpretation?

Run "/install investoday-sector-research-interpretation" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is investoday-sector-research-interpretation free?

Yes, investoday-sector-research-interpretation is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does investoday-sector-research-interpretation support?

investoday-sector-research-interpretation is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created investoday-sector-research-interpretation?

It is built and maintained by investoday (@kenneth-bro); the current version is v1.3.0.

💬 Comments