360 Secure SkillHub
360 安全技能中心

Mars Stock Analysis

核心功能:遵循标准化步骤,从业务质量、资产负债表安全等多维度对股票进行基本面评分与评级,多标的请求时完成同业排名选优,同步标注分析可信度及异常数据。典型场景:用户需分析个股、对比同业或筛选优质股票时使用。输入为股票代码,输出分维度清晰结论,仅作教育参考,不构成投资建议。

mars82311111
mars82311111
热度 35.15
最近更新 2026-05-11 17:38:40
分类:数据依赖:联网浏览搜索 / 读写文件应用场景:财务 / 市场分析 / 数据处理 / 报告撰写

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add mars-stock-analysis
Metadata
name
fundamental-stock-analysis
description
Fundamental equity analysis and peer ranking using a structured scoring playbook (quality, balance-sheet safety, cash flow, valuation, sector adjustments, confidence modifiers). Use when a user asks to analyze one or more stock tickers, compare peers, choose a best pick, or produce a fundamentals-based verdict.

fundamental-stock-analysis

  1. Read references/playbook.md before starting analysis.
  2. Follow the playbook steps exactly (input parse -> data collection -> quick screen -> scoring -> rating -> output).
  3. For multi-ticker requests, analyze each ticker first, then rank peers and select best pick with invalidation triggers.
  4. Always include confidence level and call out stale/conflicting data explicitly.
  5. Do not append any machine-readable JSON block in user-facing output.
  6. Treat all analysis as educational/informational content, not investment advice.

Security scope (clarification)

  • Use web retrieval only for ticker-relevant financial statements, filings, market/fundamental datasets, and relevant financial news.
  • Do not request, handle, or expose credentials/secrets.
  • Do not perform command execution, local file discovery unrelated to analysis, or arbitrary URL exploration outside ticker-relevant finance/news scope.

Non-goals

  • Data exfiltration or collection of private/non-public information.
  • Browser/automation tasks outside equity fundamental analysis and citation gathering.

Output discipline

  • Keep conclusions decisive and risk-aware.
  • Separate business quality, balance-sheet safety, and valuation.
  • Never fabricate missing metrics; mark NA.