← ClaudeAtlas

huashu-researchlisted

Structured web research workflow that ensures research findings are saved incrementally to files and not lost due to session interruptions. Use when the user says "research", "search for information", "look this up", "find out about", or "latest information".
Biraj2004/huashu-skills-english · ★ 3 · Web & Frontend · score 76
Install: claude install-skill Biraj2004/huashu-skills-english
# Research Skill A structured web research workflow with one core objective: real-time persistence of research findings to prevent loss due to session interruptions. ## When to Use - Conducting preliminary research before writing an article - Learning about new products, new technology, or recent releases - Gathering competitor information or industry trends - Any information-gathering task requiring multiple web searches ## Execution Workflow ### Step 1: Create the Research File Immediately - Create the file **before** starting any searches - Path: `_knowledge_base/research-<topic>-<YYYYMMDD>.md` - Initial content includes: research objective, key questions, expected output ```markdown # [Topic] Research Notes Research Date: YYYY-MM-DD Research Objective: [One-sentence summary] ## Key Questions 1. [Question 1] 2. [Question 2] 3. [Question 3] ## Findings (To be filled in progressively during research) ## Source List (Append after each search) ``` ### Step 2: Search and Save Incrementally - After every web search, immediately append findings to the file - Tag each finding with the source URL and date - Follow the source priority rules (see SHARED-RULES.md) ### Step 3: Stage Summaries - After every 3 searches, save a "stage summary" to the file - Format: `### Stage Summary (Round N)` + current key findings ### Step 4: Final Brief At the end of research, organise the file into a structured brief: ```markdown ## Research Conclusions ### Key Facts 1. [Fact 1]