Profile My skills Saved Collections Edit profile Submit a skill

← all publishers

b-m-capital-research

@b-m-capital-research source repo

36 published skills

  1. ▌
    Moat Analysis · b-m-capital-research
    公司侧壁垒的取证与判定——护城河来源分型、每一型对应的当轮可取数证据、护城河到估值倍数上下沿的传导，以及护城河正在被侵蚀的观察信号
    0
    installs
  2. ▌
    Missed Digest Items · b-m-capital-research
    查回 digest 没推但用户可能想看的事件(数量上限截断 / 同 ticker 冷却 / 噪音过滤等)
    0
    installs
  3. ▌
    Hone Administrator · b-m-capital-research
    Admin-only skill for viewing or modifying Hone source code and configuration, and for performing restarts. Available only to users listed in config.yaml admins
    0
    installs
  4. ▌
    Hari Invest · b-m-capital-research bundle
    HONE 的美股投资判断与自然对话 Skill。用于股票分析、公司基本面、护城河、财报、估值、行业、板块、宏观、ETF、持仓、仓位、评级、红绿灯、能买吗、该不该买、要不要卖、加仓、减仓、回调去留等问题。结合最新可核实证据，先给机会区、持有区、风险区或数据不足的明确研究结论，再解释老王框架、短中长期、最强反方和改变判断的条件；不冒充老王，不编造实时数据或精确交易指令。
    0
    installs
  5. ▌
    Etf Analysis · b-m-capital-research
    股票型 ETF 的可复算估值分析——按持仓权重把前十成分股的合理估值区间加权汇总成 ETF 层面的合理价区间，并披露覆盖率。用于"这只 ETF 现在能不能买 / 贵不贵 / 适不适合建仓"这类需要算出数的问题
    0
    installs
  6. ▌
    Fundamentals · b-m-capital-research
    基本面这一维的取数与判定——用分部收入回答钱从哪来，用利润率与现金流的趋势回答赚得怎么样，用净现金与偿债指标回答撑不撑得住，用增长来源回答还能不能持续。每一项都要落到"这家公司现在处于什么状态"，不是把财报数字抄一遍
    0
    installs
  7. ▌
    Industry Map · b-m-capital-research bundle
    AI 数据中心的行业树与每一行的底层估值逻辑——从 AI 侧可观测量到这一行收入/价格的传导链、该用哪族倍数、这一行最常见的估值错法、以及最该盯的那几个先行指标。个股分析命中树里的公司时，用它把「这家公司」放回「这一行的量价驱动」里，而不是每家都从头讲一遍 AI 故事。
    0
    installs
  8. ▌
    Gold Analysis · b-m-capital-research bundle
    Analyze gold, gold-linked ETFs, and gold miners through macro drivers, positioning, and event risk
    0
    installs
  9. ▌
    Skill Management · b-m-capital-research bundle
    Design, inspect, create, update, and validate Hone skills, including Codex-native discovery through workspace .agents/skills and compatibility with legacy Hone runners.
    0
    installs
  10. ▌
    Scheduled Task Management · b-m-capital-research
    Register, update, cancel, and enrich user scheduled push tasks, including event-driven reminders for portfolio holdings
    0
    installs
  11. ▌
    Stock Research · b-m-capital-research
    Canonical Hone security-research skill covering company and ETF/fund analysis, valuation framing, and criteria-based screening
    0
    installs
  12. ▌
    Market Analysis · b-m-capital-research
    Macro, policy, and market/sector analysis plus price-move attribution, grounded in official data sources and graded causal evidence
    0
    installs
  13. ▌
    Position Advice · b-m-capital-research
    Cost-basis-aware position adjustment advice — add / hold / trim / cut-loss judgments grounded in the user's actual holdings, current-turn quotes, and per-ticker event-risk checks
    0
    installs
  14. ▌
    Sector To Stock · b-m-capital-research
    把板块、主题或"帮我推荐点什么"这类宽泛诉求收敛成一份可执行的荐股清单——板块稀缺度与差异化打分排序、标的对比表（基本面/护城河/财务/估值）、按杠铃原则给出的仓位比例区间
    0
    installs
  15. ▌
    Valuation Audit · b-m-capital-research
    Auditable valuation and entry-point methodology - reconcile every input line by line, pick methods by business model, run scenarios plus reverse valuation, close with a four-state verdict
    0
    installs
  16. ▌
    Analyst Coverage · b-m-capital-research
    把投行评级、目标价与研报覆盖单独做成一层：先用带日期的评级动作流摆出谁在什么时候把评级或目标价改到多少，再用带绝对日期的检索补那条研报的论点，最后把外部锚与自算区间对照并说明分歧来自哪个输入
    0
    installs
  17. ▌
    Company Portrait · b-m-capital-research bundle
    维护长期公司画像、投资主线 与事件时间线；当用户想建立公司档案、在财报/公告后更新长期判断，或希望保留结论、为什么成立、关键证据与研究路径时使用
    0
    installs
  18. ▌
    Earnings Readout · b-m-capital-research
    财报这一维的执行口径——发布前把预期差拆成一致预期、公司指引、产业链读数与买方定价四个来源分别取数，发布后把 Beat/Miss 的幅度与决定估值中枢的变量分开写，并给出电话会原文受限时管理层措辞的落笔方式
    0
    installs
  19. ▌
    First Principles · b-m-capital-research
    把"受益于 AI"这类叙事压缩成一条可算、可证伪的公式——需求驱动量 × 单位用量 × 单价 → 供给约束 → 价格与量的走向 → 这家公司拿到多少收入与利润，再用本轮数据验证或推翻它
    0
    installs
  20. ▌
    Image Generation · b-m-capital-research
    Generate dynamic portfolio screenshots, stock-analysis infographics, or general-style images. Real data must be fetched first
    0
    installs
  21. ▌
    Influencer Views · b-m-capital-research
    把大V速报（Serenity/白毛 @aleabitoreddit、SemiAnalysis 等注册作者近期公开推文与文章，含中文翻译与 HONE 观点整理）作为「观点线索」接进个股与产业问答：用户问某只股票最近的新闻、消息、市场怎么看，或点名 Serenity/白毛/大V 时，先取 influencer_views，再把带作者、日期与原链的观点写成独立一段，与已核验事实分层。
    0
    installs
  22. ▌
    Options Analysis · b-m-capital-research
    Options and derivatives position analysis covering contract restatement, exercise-style defaults, coverage math, expiration mechanics, and leveraged/inverse ETF path decay
    0
    installs
  23. ▌
    Earnings Research · b-m-capital-research bundle
    Run the administrator-only HONE 财报前瞻 or 财报分析 workflow with current financial data, targeted web search, automatically retrieved filings and calls, the original BamangResearch/Dify analysis prompts, and a shareable watermarked PDF. Use only for the structured 财报前瞻 and 财报分析 chat entries.
    0
    installs
  24. ▌
    PDF Understanding · b-m-capital-research
    Read and analyze user-uploaded PDFs, preferably using the system-extracted text fragments, and output summaries, key points, and risk notes
    0
    installs
  25. ▌
    Chart Visualization · b-m-capital-research bundle
    把已有数字渲染成 PNG 图表（趋势/对比/结构占比/分布），当纯文字讲不清楚、或用户直接要图时使用；渲染器支持 line/area/bar/scatter/histogram/horizontal_bar，饼图用 horizontal_bar 顶替
    0
    installs
  26. ▌
    Deep Stock Research · b-m-capital-research
    Admin-only skill for launching deep research on a specific company for about 1-2 hours while reporting progress every minute
    0
    installs
  27. ▌
    Image Understanding · b-m-capital-research
    Analyze images sent by the user, such as portfolio screenshots or candlestick charts, and combine the result with tools for a fuller assessment
    0
    installs
  28. ▌
    Alipay Aipay · b-m-capital-research bundle
    支付宝AI支付站点官方 Skill。当用户选择按量付费、网站支付、APP支付进行集成、签约入驻或全流程同时完成两者时，使用此 Skill。 触发关键词："AI支付"、"接支付宝"、"接个支付"、"402收款"、"按量付费"、"快捷收款"、"快捷收单"、"网站支付"、"APP支付"、"商家入驻"、"签约入驻"等。
    0
    installs
  29. ▌
    Portfolio Management · b-m-capital-research
    Manage user holdings and watchlist (add/update/remove/watch/unwatch with ticker validation), and answer portfolio-level screening, ranking, and multi-ticker batch questions with reproducible scorecards
    0
    installs
  30. ▌
    Formal Release · b-m-capital-research bundle
    为 honeclaw 仓库执行正式版本发布：补版本号、编写 `docs/releases/vX.Y.Z.md`、完成必要验证、提交变更、推送 `main`、创建并推送 `v*` tag。用户明确要求“发正式版本 / 打 tag / release / release note / 把上次 release 到现在的变更发出去”时使用；不适用于普通 push。
    0
    installs
  31. ▌
    Company Thesis Ratings · b-m-capital-research bundle
    基于 HONE 已授权的历史演讲与研究文件所形成的公司研究卡，回答覆盖美股公司的基本面、商业模式、护城河、估值框架、风险、证伪条件与每日评分。用户询问覆盖公司的投资逻辑、公司质量、护城河、当前估值、财报变化、红黄绿灯、评分原因或多公司比较时必须优先使用；历史研究只作为判断基线，最新行情、财报、新闻、订单和估值输入必须继续通过当前证据工具核验。
    0
    installs
  32. ▌
    Notification Preferences · b-m-capital-research
    管理当前用户的市场事件推送偏好,把中文自然语言映射到 notification_prefs 工具
    0
    installs
  33. ▌
    Scarcity Differentiation · b-m-capital-research
    稀缺度与差异化的 1–5 打分方法——需求增速对供给释放定稀缺、价值为什么落到这家定差异化，每个分数挂一条当轮证据，并把两个分数换算成倍数相对锚的调整方向与量级
    0
    installs
  34. ▌
    Company Latest Developments · b-m-capital-research
    Build a dated, first-party-sourced timeline for "某公司/某项目最新进展、最近动态、最近发生了什么" questions by combining SEC filings, official press releases, dated analyst actions, ticker news, and targeted web search
    0
    installs
  35. ▌
    Event Engine Push Review · b-m-capital-research bundle
    Use in the honeclaw repository when Codex needs to review event-engine push quality for one or more actors: what each user/channel received, whether push timing and digest grouping were reasonable, whether messages were duplicated or useless, and whether useful events were filtered, omitted, capped, cooled down, or left undelivered.
    0
    installs
  36. ▌
    Event Engine Baseline Testing · b-m-capital-research bundle
    Use in the honeclaw repository when Codex needs to validate event-engine notification quality changes, run the event-engine/core/web-api regression subset, rerun the saved news classifier baseline against OpenRouter models, or add new real-news samples to the baseline fixture without overfitting.
    0
    installs