literature-review-agent

Discover and verify candidate papers via Semantic Scholar, then draft LaTeX Introduction and Related Work sections.

2|1|Updated Apr 19, 2026
One-click install
npx skills add https://github.com/raja21068/AutoResearch --skill literature-review-agent-raja21068
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: literature-review-agent
Source: https://github.com/raja21068/AutoResearch/tree/main/skills/paper-pipeline/literature-review-agent
Command: npx skills add https://github.com/raja21068/AutoResearch --skill literature-review-agent-raja21068

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python-Levenshtein, and includes scripts (resource) and references (resource) components.

What problem does it solve?

It solves the problem of producing an Introduction and Related Work section with trustworthy, execution-grounded citations rather than unverified or citation-inflated claims.

Core Features & Use Cases

  • Verified literature discovery: Runs a two-phase pipeline that first discovers candidate papers via search and then verifies them against Semantic Scholar (fuzzy title match, cutoff compliance, deduplication).
  • Citation pool hardening: Produces a verified citation_pool.json plus a deterministic refs.bib, enforcing strict temporal and integration rules for the writing step.
  • Writing with integration guarantees: Drafts Introduction + Related Work in LaTeX while later enforcing a ≥90% citation coverage gate against the verified pool.

Quick Start

Use the literature-review-agent skill to create a verified bibliography and draft your paper’s Introduction and Related Work from an existing workspace/outline.json plan and writing template.

Frequently Asked Questions about literature-review-agent

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

FAQPage Schema
How do I generate a verified bibliography and draft an Introduction from an outline?

To generate a verified bibliography and draft an Introduction, discover candidate papers, verify them via Semantic Scholar fuzzy matching with strict temporal cutoff, and draft LaTeX sections using only the verified citation pool to produce citation_pool.json and refs.bib.

How does Semantic Scholar fuzzy matching verify academic citations?

Semantic Scholar fuzzy matching verifies academic citations by comparing candidate paper titles against the database, applying strict temporal cutoff and deduplication rules to ensure only execution-grounded, verified references enter the citation pool.

What is the best way to enforce citation integration in academic writing?

The best way to enforce citation integration in academic writing is to draft Introduction and Related Work sections using canonical BibTeX keys, applying a strict coverage gate that ensures at least 90% of citations from the verified pool are integrated.

Do I need a workspace outline and conference guidelines to use this literature review pipeline?

Yes, this literature review pipeline requires an existing workspace outline and conference guidelines to derive the temporal cutoff, ensuring the generated bibliography and drafted sections comply with specific academic submission deadlines.

Can I produce BibTeX files and LaTeX drafts from unverified literature search results?

No, producing BibTeX files and LaTeX drafts from unverified results is restricted. The pipeline hardens citations into a deterministic refs.bib and drafts LaTeX sections exclusively from the verified citation pool to prevent citation inflation.

Why does my literature review require deduplication and a temporal cutoff?

Your literature review requires deduplication and a temporal cutoff to prevent redundant entries and ensure all verified citations comply with conference deadlines, producing a trustworthy, execution-grounded bibliography for academic writing.