Agent Skills

Conduct a systematic literature review on an academic topic. Use when the user asks for a literature review, survey, or systematic overview of a research area.

Install

npx skills add https://github.com/collaborative-deep-research/agent-papers-cli --skill literature-review
SKILL.md

Conduct a systematic literature review on "$ARGUMENTS" using the paper and paper-search CLI tools.

1. Define Scope

Before searching, clarify with the user:

  • Topic boundaries and key terms
  • Year range (default: last 5 years)
  • Target venues or communities (if any)
  • Desired number of papers (default: 15-20 core papers)

2. Multi-Query Search

Search with multiple query variations to maximize coverage:

paper-search semanticscholar papers "<main query>" --limit 20 --year <range>
paper-search semanticscholar papers "<synonym query>" --limit 20 --year <range>
paper-search semanticscholar papers "<related query>" --limit 20 --year <range>
paper-search google scholar "<topic>"

Deduplicate results by title/paper ID.

3. Triage

For each unique paper found:

paper-search semanticscholar details <paper_id>
paper skim <arxiv_id> --lines 2

Categorize as: highly relevant / somewhat relevant / not relevant.

4. Deep Analysis

For highly relevant papers:

paper outline <arxiv_id>
paper read <arxiv_id> introduction
paper read <arxiv_id> method
paper read <arxiv_id> results
paper read <arxiv_id> conclusion

Take structured notes on each paper: problem, method, key results, limitations.

5. Citation Graph Exploration

For seminal papers, find related work:

paper-search semanticscholar citations <paper_id> --limit 20
paper-search semanticscholar references <paper_id> --limit 20

Add any important papers discovered this way back to the triage step.

6. Produce Report

Organize findings by theme, not by paper. Include:

  • Overview of the field and its evolution
  • Key methods and approaches (with comparisons)
  • Main results and findings
  • Open questions and future directions
  • Complete reference list with paper IDs and URLs
  • BibTeX entries for all cited papers (use paper bibtex <arxiv_id> to generate)

Guidelines

  • Aim for breadth first: cover all major approaches before going deep on any one.
  • Note citation counts and venues to gauge paper impact.
  • Flag contradictory findings explicitly.
  • Distinguish between empirical results and theoretical claims.

Related skills

researchmattpocock575KInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.paper-context-resolverlllllllama451KRigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing Renv-and-assets-bootstraplllllllama450KRigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.ai-research-explorelllllllama311KRigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow c

Search skills and MCP servers

Fuzzy search across 23,137 skills and servers