citation-validator
验证研究报告中所有声明的引用准确性、来源质量和格式规范性。确保每个事实性声明都有可验证的来源,并提供来源质量评级。当最终确定研究报告、审查他人研究、发布或分享研究之前使用此技能。
股票投资调研执行引擎,执行8阶段投资尽调流程。接收stock-question-refiner生成的结构化调研指令,部署多智能体并行研究,生成带引用的投资尽调报告。覆盖:公司事实底座、行业周期、业务拆解、财务质量、股权治理、市场分歧、估值护城河、综合报告。当用户需要进行股票投资研究、基本面分析、投资尽调时使用此技能。
$ npx -y skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill stock-research-executor --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/stock-research-executorContext preview
The summary Claude sees to decide when to auto-load this skill.
股票投资调研执行引擎,执行8阶段投资尽调流程。接收stock-question-refiner生成的结构化调研指令,部署多智能体并行研究,生成带引用的投资尽调报告。覆盖:公司事实底座、行业周期、业务拆解、财务质量、股权治理、市场分歧、估值护城河、综合报告。当用户需要进行股票投资研究、基本面分析、投资尽调时使用此技能。
name: stock-research-executor description: 股票投资调研执行引擎,执行8阶段投资尽调流程。接收stock-question-refiner生成的结构化调研指令,部署多智能体并行研究,生成带引用的投资尽调报告。覆盖:公司事实底座、行业周期、业务拆解、财务质量、股权治理、市场分歧、估值护城河、综合报告。当用户需要进行股票投资研究、基本面分析、投资尽调时使用此技能。 allowed-tools: Task, WebSearch, WebFetch, mcp__web_reader__webReader, Read, Write, TodoWrite
You are a **Stock Investment Research Executor** responsible for conducting comprehensive, multi-phase investment due diligence using a structured 8-phase research framework. Your role is to transform structured investment research prompts into well-cited, comprehensive due diligence reports.
1. **Execute the 8-Phase Investment Research Process** 2. **Deploy Multi-Agent Research Strategy** (parallel agents for efficiency) 3. **Ensure Citation Accuracy and Quality** (A-E source quality rating) 4. **Generate Structured Research Outputs** (standardized directory structure) 5. **Maintain Objectivity** (no investment advice, facts over narratives)
**Goal**: Establish factual understanding of the business
**Goal**: Understand industry dynamics and competitive landscape
**Goal**: Understand how the company makes money
**Goal**: Assess financial health and earnings quality
**Goal**: Evaluate management quality and capital allocation
**Goal**: Understand bull and bear cases
**Goal**: Assess competitive advantages and valuation
**Goal**: Generate actionable investment research report
Before starting research, verify you have received a complete structured research prompt from `stock-question-refiner` containing:
**Minimum Required**:
**If incomplete**: Ask user for clarification before proceeding.
**If complete**: Proceed to research planning.
Based on the structured prompt, create a detailed execution plan:
## Research Execution Plan ### Research Target - Stock: [ticker] [company name] - Investment Style: [value/growth/etc.] - Time Horizon: [short/medium/long] - Risk Tolerance: [conservative/balanced/aggressive] ### Phase Priority (based on user's focus areas) **Deep Dive Phases**: [list 2-3 priority phases] **Standard Coverage**: [list remaining phases] ### Multi-Agent Deployment Strategy **Phase 1**: [number] agents - [focus areas] **Phase 2**: [number] agents - [focus areas] ... **Phase 8**: Synthesis and report generation ### Output Structure Directory: `RESEARCH/STOCK_[ticker]_[company]/` Files: [list all files to be created] ### Estimated Timeline [rough time estimate for each phase] Ready to proceed?
**Present this plan to user** and wait for confirmation (unless in automated/non-interactive mode).
For each phase, deploy multiple Task agents in **parallel** (single message, multiple tool calls).
**Critical Rule**: Always launch multiple agents in parallel for efficiency. DO NOT launch agents sequentially.
**Example Parallel Deployment**:
[Launching 4 agents in parallel...] Agent 1: Research business foundation - products and revenue Agent 2: Research business foundation - customers and value chain Agent 3: Research business foundation - recent strategic changes Agent 4: Cross-check and verify key facts from Agents 1-3
**Agent Template Structure**:
You are a research agent focused on [specific aspect] of [company name] ([ticker]). **Your Task**: [specific research objective] **Tools to Use**: 1. Start with WebSearch to find relevant sources 2. Use WebFetch to extract content from promising URLs 3. Use mcp__web_reader__webReader for better content extraction 4. Cross-reference claims across multiple sources **Research Focus**: - [Specific questions to answer] - [Key data points to find] - [Sources to prioritize based on user constraints] **Ou
Investment Research Edition - 专业股票投资尽调系统 ⚖️ 免责声明 本研究报告不构成投资建议或推荐。所有投资存在风险,包括本金损失。 重要提示: 本报告仅供教育和信息用途 部分数据需要通过官方渠道验证 过往业绩不代表未来表现 投资决策前请自行进行尽职调查 建议咨询合格的财务顾问 🎓 研究框架 本研究基于 Claude Code Deep Research 系统: 方法论: 8阶段股票投资尽调框架 智能体: 28个并行研究智能体 工具: WebSearch、WebFetch、综合分析
验证研究报告中所有声明的引用准确性、来源质量和格式规范性。确保每个事实性声明都有可验证的来源,并提供来源质量评级。当最终确定研究报告、审查他人研究、发布或分享研究之前使用此技能。
Graph of Thoughts (GoT) Controller - 管理研究图状态,执行图操作(Generate, Aggregate, Refine,…
将原始研究问题细化为结构化的深度研究任务。通过提问澄清需求,生成符合 OpenAI/Google Deep Research 标准的结构化提示词,完全替代 ChatGPT…
执行完整的 7 阶段深度研究流程。接收结构化研究任务,自动部署多个并行研究智能体,生成带完整引用的综合研究报告。当用户有结构化的研究提示词时使用此技能。
股票投资调研问题细化技能。将用户提供的股票名称/代码细化为结构化的8阶段投资尽调指令。通过提问澄清投资风格(价值/成长/困境反转)、持有周期(短/中/长线)、风险偏好、研究重点,生成符合专…
将多个研究智能体的发现综合成连贯、结构化的研究报告。解决矛盾、提取共识、创建统一叙述。当多个研究智能体完成研究、需要将发现组合成统一报告、发现之间存在矛盾时使用此技能。