/survey

Generate paper-ready Related Work sections from wiki knowledge with verified citations.

1.6k|208|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/skyllwt/AutoSci --skill survey-skyllwt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: /survey
Source: https://github.com/skyllwt/AutoSci/tree/main/.claude/skills/survey
Command: npx skills add https://github.com/skyllwt/AutoSci --skill survey-skyllwt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires tools/research_wiki.py, tools/fetch_s2.py.

What problem does it solve?

It turns existing wiki research knowledge into a coherent, citation-disciplined Related Work section that can be inserted directly into an academic paper.

Core Features & Use Cases

  • Thematic grouping to narrative flow: Organizes cited work by research direction (not paper-by-paper) and writes paragraphs that follow academic Related Work conventions.
  • Wiki-grounded evidence and citation discipline: Selects candidate papers from wiki/papers/, then verifies citation keys according to the repo’s citation-verification rules, avoiding fabricated references.
  • Multi-format output with archiving: Produces terminal-ready text in either LaTeX or Markdown citation styles and archives the generated section under wiki/outputs/related-work-{slug}-{date}.md, updating graph edges for provenance.

Quick Start

Ask the skill to write the Related Work for a topic by running /survey with your research question as input and requesting LaTeX output.

Frequently Asked Questions about /survey

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

FAQPage Schema
How do I generate a related work section from my research notes without fabricating citations?

Generating a related work section without fabricated citations involves transforming wiki knowledge into theme-based narrative while enforcing citation-verification rules against existing papers, preventing fabricated references in academic writing.

How do I organize my literature review by research direction instead of paper-by-paper?

Organizing a literature review by research direction involves grouping candidate papers from your wiki into 3-8 theme-based sections, creating a narrative flow that follows academic writing conventions rather than a paper-by-paper summary.

Can I output my related work section in LaTeX format directly from my wiki?

Yes, you can output your related work section in LaTeX format directly from your wiki. The skill supports parameterizable prompt selection for either LaTeX or Markdown citation styles from your synthesized wiki knowledge.

Does the related work generation tool require a specific wiki structure to function?

Yes, related work generation requires reading and synthesizing specific wiki structures including wiki/papers, wiki/concepts, and wiki/topics, along with graph context, to group content into direction-based sections accurately.

How do I archive generated related work sections and track provenance in my research pipeline?

To archive generated related work sections and track provenance, the skill saves output under wiki/outputs/related-work-{slug}-{date}.md and optionally adds derived_from graph edges to maintain research pipeline provenance.

What is the best way to write a paper-ready related work section from existing research notes?

The best way to write a paper-ready related work section from existing research notes is using a PAPER_PLAN-based template that applies idea-slug driven organization to synthesize wiki context into citation-disciplined academic narrative.