← 返回 Skills 市场
hypier

equity-research-analyst-skill

作者 hypier · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
47
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install equity-research-analyst
功能描述
Use when the user asks for institutional-style equity research on a public company, ETF, or TradingView-resolvable ticker, including initiation reports, earn...
使用说明 (SKILL.md)

Equity Research Analyst

Produce institutional-grade equity research deliverables covering nine distinct workflows. Each workflow has its own dedicated reference file in references/workflows/; this SKILL.md is the dispatcher and the hub for cross-workflow conventions.

Loading Strategy

Keep context tight and load only the files needed for the active task.

  1. Match the request to exactly one workflow in references/workflows/.
  2. Read only that workflow file first.
  3. Open deep-dive references only when the active workflow points to them.
  4. Read references/tradingviewapi.md before opening anything in references/tradingviewapi-docs/.
  5. Treat references/tradingviewapi-docs/ as a lookup bundle: search by endpoint name or JSON field, then open the smallest relevant file instead of loading an entire large example file.

tradingviewapi lookup guide

  • Start with references/tradingviewapi.md for task-to-endpoint mapping.
  • Use references/tradingviewapi-docs/README.md for file selection inside the bundled API docs.
  • Use references/tradingviewapi-docs/openapi.json as the source of truth for exact parameters, defaults, and allowed enum values.
  • Use references/tradingviewapi-docs/examples/ mainly for concrete URL patterns and response shapes, not for deciding which asset-class parameter should be used.
  • If an example and the OpenAPI spec disagree, trust openapi.json for parameters and trust executed examples only for payload shape after the request has been validated against the spec.
  • Search patterns that usually find the right example quickly:
    • GET /api/market-data/{symbol}
    • GET /api/calendar/earnings
    • GET /api/search/market/{query}
    • earnings_release_next_date
    • analyst-recommendations

When to invoke which workflow

Match the user request to one of the nine workflows below and read the corresponding reference file from references/workflows/. If the request is ambiguous, ask the user to confirm before committing.

Workflow Trigger phrases Reference
Initiating coverage "initiation report", "first-time coverage", "new coverage on X", "write a full report on X" references/workflows/initiating-coverage.md
Earnings analysis "earnings update", "Q1/Q2/Q3/Q4 results for X", "post-earnings report", "beat/miss analysis" references/workflows/earnings-analysis.md
Earnings preview "earnings preview", "what to watch in X's earnings", "pre-earnings note" references/workflows/earnings-preview.md
Catalyst calendar "catalyst calendar", "upcoming events", "earnings calendar", "event tracker" references/workflows/catalyst-calendar.md
Morning note "morning note", "daily brief", "pre-market note", "morning wrap" references/workflows/morning-note.md
Sector overview "sector report", "industry overview", "peer comp", "sector deep-dive" references/workflows/sector-overview.md
Thesis tracker "thesis tracker", "investment thesis review", "revisit thesis on X", "thesis check" references/workflows/thesis-tracker.md
Model update "update the model", "model maintenance", "refresh estimates", "re-forecast" references/workflows/model-update.md
Idea generation "screen for X", "find me stocks that", "idea generation", "investment ideas" references/workflows/idea-generation.md

Single-workflow discipline (especially for initiating-coverage)

The initiation-report workflow has five sequential tasks (Company Research → Financial Modeling → Valuation → Charts → Assembly). Execute one task per user request, verify prerequisites before the next task, and never auto-chain. Details in references/workflows/initiating-coverage.md.

Primary data source: tradingviewapi

Before Web Search, pull structured numeric data (financials, TTM ratios, analyst consensus, calendars, prices, technicals, news) from the bundled tradingviewapi (TradingView proxy). One call to /api/market-data/{symbol} covers ~70% of the numeric content of a typical research report.

  • Full endpoint map, curl examples, and JSON-path-to-report-field mapping: references/tradingviewapi.md
  • OpenAPI spec and example responses: references/tradingviewapi-docs/

Authentication: users must provide their own RAPIDAPI_KEY env var (RapidAPI hosted).

Use Web Search ONLY for narrative content: MD&A text, forward guidance wording, earnings call transcripts, segment breakdowns, risk factors, management bios, industry research, FDA/regulatory decisions. Pull raw SEC 10-K/10-Q only when direct quotation or audit is required.

Global conventions

Data freshness

  • Training data is outdated. Before writing, verify today's date and confirm that the latest fetched data is current (next earnings date, last reported quarter).
  • If data.current.fiscal_period_current is >90 days old, flag as "last reported" and Web Search for newer disclosures.
  • Treat data.current.fiscal_period_current as a structured provider label, not the final narrative quarter label. If company IR / SEC naming differs (for example provider 2025-Q4 vs company-reported Fiscal 2026 Q4), use the latest primary-source wording in the written report and cite that source explicitly.

Ticker resolution

  • Always use EXCHANGE:TICKER format (e.g., NASDAQ:AAPL, NYSE:JPM).
  • If the user supplies only a company name, resolve it via /api/search/market/{query}?filter=stock before proceeding.
  • Use the resolved symbol from /api/search/market/{query} as the canonical identifier for the workflow. Do not rely on data.company.ticker or data.company.exchange from /api/market-data/{symbol} as the canonical listing identifier, because those fields may be null or may reflect a quote venue rather than the primary exchange.

Citation standard

Every numeric fact in a deliverable must cite its source:

Source: Structured data via tradingviewapi (TradingView); fetched [YYYY-MM-DD]
        Endpoint: /api/market-data/NASDAQ:AAPL
        Fiscal period: 2026-Q1

SEC filings keep separate EDGAR hyperlinks. When consensus data comes from the API, cite the analyst-recommendations endpoint explicitly instead of a generic terminal label.

Output formatting

  • Default font for Word deliverables: Times New Roman.
  • Do not add emojis to research reports unless the user explicitly requests them.
  • Follow the page/structure templates in assets/initiating-coverage/report-template.md for initiation reports.
  • Use references/initiating-coverage/quality-checklist.md for final initiation-report QA.
  • Other workflows use the template embedded in their own reference file.

No shortcuts

  • Deliver exactly the outputs specified in each workflow reference. Do not create extra "completion summaries", "executive summaries", or "quick reference guides".
  • Do not fabricate data. Missing field → "N/A". Missing consensus → state "consensus not available".

Fallback strategy

  1. tradingviewapi unavailable → fall back to Web Search + SEC EDGAR per the workflow reference.
  2. Ticker unresolved → ask the user for EXCHANGE:TICKER.
  3. Ambiguous workflow → ask the user which deliverable they want.

Directory layout

equity-research-analyst/
├── SKILL.md                              # This file
├── references/
│   ├── tradingviewapi.md                   # Endpoint map + report-field mapping
│   ├── tradingviewapi-docs/                # Bundled API spec + examples (snapshot)
│   │   ├── README.md
│   │   ├── openapi.json
│   │   └── examples/ (12 .md files)
│   ├── workflows/                        # One dispatch target per workflow
│   │   ├── initiating-coverage.md
│   │   ├── earnings-analysis.md
│   │   ├── earnings-preview.md
│   │   ├── catalyst-calendar.md
│   │   ├── morning-note.md
│   │   ├── sector-overview.md
│   │   ├── thesis-tracker.md
│   │   ├── model-update.md
│   │   └── idea-generation.md
│   ├── earnings-analysis/                # Deep-dive references for earnings workflow
│   │   ├── workflow.md
│   │   ├── report-structure.md
│   │   └── best-practices.md
│   └── initiating-coverage/              # Deep-dive references for initiation tasks 1-5
│       ├── task1-company-research.md
│       ├── task2-financial-modeling.md
│       ├── task3-valuation.md
│       ├── task4-chart-generation.md
│       ├── task5-report-assembly.md
│       ├── quality-checklist.md
│       └── valuation-methodologies.md
├── assets/
│   └── initiating-coverage/              # Templates used in output
│       └── report-template.md
安全使用建议
This skill appears to genuinely implement institutional equity‑research workflows, but there are important mismatches you should resolve before installing or enabling it. Key actions to take: 1) Ask the publisher to explicitly declare required credentials and runtime tools (RAPIDAPI_KEY, Python and which packages, and that the DOCX/XLS skills or toolchain are required). 2) Do not supply your RapidAPI key until you confirm where and how it will be stored/used; prefer per-agent scoped keys and least privilege. 3) Confirm the environment where the agent will run has the required tooling (Python, matplotlib/plotly, an Office-generation capability) or be prepared for failures. 4) If you cannot validate the publisher, run the skill in a sandboxed agent/container with no access to high-value credentials or internal network resources. 5) If you need assurance, request a minimal test mode: a run that performs only symbol-resolution and a read-only TradingView API call (no DOCX/XLS generation) so you can validate behavior and network endpoints. These steps will reduce the risk posed by the undeclared API key and missing dependency information.
功能分析
Type: OpenClaw Skill Name: equity-research-analyst Version: 1.0.0 The skill bundle is a highly professional and comprehensive framework for an AI agent to perform institutional-grade equity research. It utilizes a structured approach for data collection via the TradingView API (RapidAPI) and provides detailed workflows for financial modeling, valuation, and report assembly. The Python code provided in 'references/initiating-coverage/task4-chart-generation.md' is strictly limited to data visualization using standard libraries (matplotlib, seaborn, pandas). There is no evidence of malicious intent, data exfiltration, or prompt injection; the instructions are consistently aligned with the stated purpose of financial analysis.
能力标签
cryptocan-make-purchasesrequires-sensitive-credentials
能力评估
Purpose & Capability
The skill's name/description align with the included references and workflows (initiating coverage, earnings notes, valuation, etc.). However, the SKILL.md explicitly requires use of a RapidAPI-hosted TradingView proxy (RAPIDAPI_KEY) and external tooling (Python libraries, DOCX skill, Excel/XLS generation) that are not declared in the registry metadata (no required env vars, no required binaries, no primary credential). That mismatch is disproportionate: a research skill legitimately needs an API key and tooling, but those requirements should be declared.
Instruction Scope
SKILL.md is comprehensive and mostly scoped to research workflows and data sources (tradingviewapi for numeric data, web search for narrative). It directs creation of large deliverables (30+ page DOCX, 10k+ words, XLS model, 20–30+ PNG charts) and to produce clickable hyperlinks to external sources. The instructions do not ask the agent to read unrelated system files or harvest secrets, but they do assume access to external networks, the RapidAPI key, and local tooling to generate DOCX/XLS and images — all reasonable for the stated purpose but not declared in the packaging metadata, so the runtime behavior is under-specified.
Install Mechanism
There is no install spec (instruction-only), which lowers install-time risk. However, the skill expects use of Python (matplotlib/plotly), generation of many image files, and the DOCX/XLS skills or tooling to assemble deliverables. Because those runtime dependencies are not declared, an agent might attempt to install or execute tools not present on the host, or fail in unexpected ways. The absence of declared dependencies increases operational risk and obscures what will be required at runtime.
Credentials
SKILL.md explicitly says users must provide RAPIDAPI_KEY (RapidAPI) for TradingView data and references an x-rapidapi-host header and base URL, but the skill registry lists no required environment variables or primary credential. This is a direct inconsistency: the skill will function only if a secret API key is supplied, yet that key is not declared as required in the metadata. No other unrelated credentials are requested, but the missing RAPIDAPI_KEY declaration is material and increases the chance of misconfiguration or accidental key exposure.
Persistence & Privilege
The skill does not request elevated persistence (always: false) and does not claim to modify other skills or global agent settings. It requires creating output files (DOCX, XLS, PNG) as part of normal operation, which is reasonable for its purpose. No automatic autonomous invocation flagging is present beyond the platform default.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install equity-research-analyst
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /equity-research-analyst 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of the Equity Research Analyst skill. - Supports nine institutional equity research workflows, each triggered by specific user requests and dispatched to dedicated workflow files. - Integrates TradingView API as the primary data source for financials, consensus, technicals, and calendar data. - Implements strict conventions for data citation, ticker resolution, and output formatting. - Defines a clear fallback strategy for missing data or ambiguous requests. - Enforces single-workflow discipline with detailed guidance for initiation reports and deliverable standards.
元数据
Slug equity-research-analyst
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

equity-research-analyst-skill 是什么?

Use when the user asks for institutional-style equity research on a public company, ETF, or TradingView-resolvable ticker, including initiation reports, earn... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 47 次。

如何安装 equity-research-analyst-skill?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install equity-research-analyst」即可一键安装,无需额外配置。

equity-research-analyst-skill 是免费的吗?

是的,equity-research-analyst-skill 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

equity-research-analyst-skill 支持哪些平台?

equity-research-analyst-skill 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 equity-research-analyst-skill?

由 hypier(@hypier)开发并维护,当前版本 v1.0.0。

💬 留言讨论