topic_survey

Coordinate local vec-db, Semantic Scholar, and web searches to map research topics.

2|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/Gonglitian/agent-skills --skill topic-survey
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: topic_survey
Source: https://github.com/Gonglitian/agent-skills/tree/main/skills/topic_survey
Command: npx skills add https://github.com/Gonglitian/agent-skills --skill topic-survey

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quickly map a research field, identify sub-topics and key papers, and generate a structured literature review with a clear reading plan.

Core Features & Use Cases

  • Interactive scoping to clarify the core question, scope, seeds, and depth for a literature survey.
  • Multi-source discovery that coordinates local vec-db searches, Semantic Scholar, and web results to surface representative papers.
  • Structured output that organizes findings into sub-topics, key papers, and actionable next steps for deeper reading.

Quick Start

Provide a topic and optional seeds to start the survey, and the Skill will generate an initial landscape and a plan for deeper exploration.

Frequently Asked Questions about topic_survey

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I map a research field and identify key papers for a literature review?

To map a research field for a literature review, this Skill coordinates searches across local vec-db, Semantic Scholar, and web sources to locate key papers and sub-topics. It outputs a structured literature map with seeds and follow-up steps for deeper synthesis.

What is the best way to explore sub-topics when starting a literature survey?

The best way to explore sub-topics in a literature survey is through interactive scoping. This feature clarifies your core question, scope, and depth, then organizes multi-source discovery results into sub-topics with actionable reading plans for guided exploration.

How do I use Semantic Scholar and local vector databases together for topic exploration?

You can use Semantic Scholar and local vec-db together by providing a research topic and optional seeds. The Skill coordinates these sources automatically, running multi-source discovery to surface representative papers and assemble a unified literature map.

Can I generate a structured reading plan from an initial literature search?

Yes, you can generate a structured reading plan from an initial literature search. After mapping your research topic and identifying key papers, the Skill produces a user-guided plan and prompts designed to drive deep dives, reading lists, and final synthesis into a review.

Do I need to provide seed papers to start a literature survey?

No, you do not need to provide seed papers to start a literature survey. You can simply provide a research topic, and the Skill will generate an initial landscape using local vec-db, Semantic Scholar, and web sources, though optional seeds can refine the scoping.

What are the limitations of using automated tools for literature review synthesis?

Automated literature review tools provide a structured landscape and reading plan but require user-guided deep dives for final synthesis. The output organizes sub-topics and key papers, yet the actual reading and integration into a cohesive review remain a manual next step.