gap-analysis

name	Gap Analysis
description	This skill should be used when the user asks to "analyze research gaps", "find research gaps", "generate research ideas", "discover research opportunities", "analyze papers for gaps", or runs /ra:gap-analysis. It reads a collection of papers, identifies research gaps, and proposes novel research ideas.

Research Gap Analysis

1. Prerequisite Check

Read ${CLAUDE_PLUGIN_ROOT}/.local.md. Check if cv_profile_analyzed is true and the ## CV Profile section is populated.

If cv_profile_analyzed is not true or the ## CV Profile section is empty, output:

未检测到 CV 画像。 请先运行 /ra:cv-analyze 分析你的 CV，其他功能依赖 CV 画像数据。

Output this message and stop immediately.

2. Personalization Mode

Ask the user: "是否需要个性化定制？会问你 4-5 个问题来生成更贴合你需求的结果。"

If the user declines personalization

Use defaults based on the user's cv_profile and the papers provided. Skip to step 3.

If the user accepts personalization

Ask the following 4-5 questions using AskUserQuestion:

1. Specific research direction: "你想聚焦在哪个具体的研究方向？"
2. Technical capability scope: "你的技术能力范围是什么？" (pre-fill suggested from cv_profile technical skills)
3. Contribution type preference: "你更偏好理论贡献还是应用贡献？" (theoretical / applied / both)
4. Existing ideas: "你已经有任何想法或假设吗？"
5. Target publication venue: "你的目标发表场所是哪里？" (e.g., CHI, SIGGRAPH, NeurIPS, CVPR, UIST, CSCW)

3. Paper Source Selection

First, ask the user whether they want to use Zotero:

"是否使用 Zotero 获取论文？（需要安装 Zotero 桌面端并配置 MCP 插件）"

If the user chooses NOT to use Zotero

Only offer the folder path option:

Ask the user to provide a folder path containing PDF files. Then:

1. Use Glob to find all PDF files in the directory: **/*.pdf
2. List the found files and confirm with the user
3. Use each file path as input to the paper-reader agent

If the user chooses to use Zotero

Offer both options:

Ask: "请提供论文来源：1. 文件夹路径（包含 PDF 文件的目录）2. Zotero 集合名称"

Option A: Folder Path — same as above

Option B: Zotero Collection

1. Use zotero_get_collections to find the collection key
2. Use zotero_get_collection_items with the collection key to list papers
3. Display the list and confirm with the user
4. Use each item key as input to the paper-reader agent

4. Parallel Paper Processing

Launch paper-reader Agent for each paper (up to 5 in parallel). Send a single message with multiple Agent tool calls, one per paper.

For each paper, pass:

• The paper identifier (file path or Zotero item key)
• Instruction to return the structured summary

Each paper-reader agent returns:

### [Paper Title]
**Authors**: [author list]
**Venue**: [conference/journal], **Year**: [year]

**Core Theoretical Contributions**:
- [contribution 1]
- [contribution 2]

**Core Experimental Contributions**:
- [experiment 1]
- [experiment 2]

**Methodology Significance**:
[Why is their approach important or novel?]

**Limitations**:
- [limitation 1]
- [limitation 2]

Processing Strategy

• If there are more than 5 papers, process them in batches of 5
• Wait for each batch to complete before launching the next
• Collect all summaries before proceeding to cross-comparison

5. Cross-Comparison and Gap Identification

After collecting all paper summaries, perform a systematic cross-comparison to identify research gaps in four categories:

5.1 Methodological Gaps

• Approaches or methods not yet explored in this research area
• Combinations of existing methods that have not been tried
• New methodologies that could address limitations of current approaches
• Techniques from adjacent fields that could be transferred

5.2 Application Gaps

• Domains or contexts not yet covered by existing research
• Populations or user groups not studied
• Real-world scenarios not addressed
• Cross-cultural or cross-domain applications

5.3 Theoretical Limitations

• Unresolved questions or contradictions in the literature
• Assumptions that remain untested
• Theoretical frameworks that need extension or revision
• Foundational gaps in understanding

5.4 Technical Improvements

• Scalability limitations of current approaches
• Efficiency improvements (speed, resource usage)
• Accuracy or performance enhancements
• Robustness and generalizability improvements
• Integration or interoperability opportunities

6. Generate Research Ideas

Based on the identified gaps and the user's cv_profile, generate 3-5 research ideas. For each idea, provide:

Idea Format

### Idea [N]: [Short Title]

**Research Question**:
[Clear, specific, and answerable research question]

**Proposed Methodology**:
[Step-by-step description of the approach, methods, and techniques to be used]

**Novelty Justification**:
[How this differs from existing work — reference specific papers and explain what gap this fills]

**Feasibility Assessment**:
- Based on your technical skills: [skill match assessment]
- Required resources: [datasets, equipment, participants, compute]
- Estimated timeline: [rough time estimate]
- Feasibility score: [High / Medium / Low]

**Potential Publication Venues**:
- Primary: [venue 1]
- Secondary: [venue 2]

Feasibility Assessment Logic

Compare the required technical skills for each idea against the user's cv_profile:

• High feasibility: User has 80%+ of required skills
• Medium feasibility: User has 50-80% of required skills (identifiable skill gaps to fill)
• Low feasibility: User has <50% of required skills (significant preparation needed)

7. Present and Iterate

Initial Presentation

1. Present the gap analysis summary (organized by the four gap categories)
2. Present all 3-5 research ideas with full details
3. Ask the user which ideas they find most interesting

Iterative Refinement

Based on user feedback:

1. Refine selected ideas with more detail
2. Address concerns about feasibility
3. Suggest modifications or combinations of ideas
4. Help prioritize ideas based on user's goals and timeline

Final Output

Provide a prioritized list with:

• Recommended primary research direction
• Backup options
• Immediate next steps (literature to read, skills to develop, experiments to pilot)