Search skills/

Claude research skill options for sources and citations

22 skills and skill packs reviewed. Sorted by verdict, then depth — how much real expertise ships with the skill — not by popularity.

Our picks

last30days

mvanhorn/last30days-skill

Research skill that searches Reddit, X, YouTube, Hacker News, Polymarket and more for the last 30 days, scores results by engagement, and writes a cited brief.

Recommended Broad access Depth 5/5

Co-Researcher

poemswe/co-researcher

Academic research plugin with 15 skills; literature reviews check citations and quotes against OpenAlex, arXiv and Europe PMC.

Recommended Check first Depth 4/5

teach

skill in mattpocock/skills

Turns the current directory into a multi-session teaching workspace with a mission, resources, learning records and self-contained HTML lessons.

Recommended No issues Depth 3/5

All 22 reviewed

How we review
Skill Verdict Security Depth Stars Installs
last30days mvanhorn/last30days-skill Research skill that searches Reddit, X, YouTube, Hacker News, Polymarket and more for the last 30 days, scores results by engagement, and writes a cited brief.
Recommended Broad access 5/5 66K —
Co-Researcher poemswe/co-researcher Academic research plugin with 15 skills; literature reviews check citations and quotes against OpenAlex, arXiv and Europe PMC.
Recommended Check first 4/5 134 —
teach skill in mattpocock/skills Turns the current directory into a multi-session teaching workspace with a mission, resources, learning records and self-contained HTML lessons.
Recommended No issues 3/5 306K 883K
Journalism agent skills (jamditis) jamditis/claude-skills-journalism Plugin packages of skills for journalists and researchers (verification, FOIA, AP style) plus dev, newsroom-ops and security toolkits for Claude Code and Codex.
Recommended Check first 3/5 438 —
Family History Planning Skill emaynard/claude-family-history-research-skill Genealogy skill that plans research before searching, then covers citations, conflicting evidence and logs under the GPS.
Recommended No issues 3/5 124 —
Academic Research Skills (ARS) Imbad0202/academic-research-skills Five very detailed Claude Code skills for the academic pipeline: deep research, paper writing, peer review, revision and systematic-review screening.
With caveats Check first 5/5 54K —
notebooklm-py teng-lin/notebooklm-py Unofficial Python API, CLI and agent skill for Google's NotebookLM (now Gemini Notebook), built on undocumented endpoints.
With caveats Check first 5/5 20K —
ex-USSR Kinship Research DenisSergeevitch/ex-ussr-kinship-research An evidence-first genealogy skill for the former USSR and Russian Empire: church and civil records, archives, evidence registers, an audit script and an HTML report.
With caveats Check first 5/5 1.2K —
RigorPilot Skills lllllllama/RigorPilot-Skills Eleven skills that reproduce deep-learning repos from their README with a reviewed plan, bounded run and evidence bundle.
With caveats Check first 4/5 486 4.3M
Scientific Agent Skills K-Dense-AI/scientific-agent-skills K-Dense's library of 177 science skills covering bioinformatics, chemistry, clinical research, data analysis, lab tooling and scientific writing.
With caveats Broad access 4/5 55K —
ARIS (Auto-Research-In-Sleep) wanshuiyin/Auto-claude-code-research-in-sleep A large markdown skill set for autonomous ML research: idea discovery, experiment runs and cross-model review loops where a second model critiques the first.
With caveats Broad access 4/5 18K —
AI Research Skills (Orchestra) Orchestra-Research/AI-Research-SKILLs 98 skills for AI research and ML engineering, plus an autoresearch orchestrator that runs a two-loop experiment and synthesis cycle.
With caveats Broad access 4/5 14K —
Deep Research Skill 199-biotechnologies/claude-deep-research-skill A single research skill with an 8-phase pipeline, parallel search, source scoring, claim and citation checks, and report output in Markdown, HTML and PDF.
With caveats Check first 4/5 1.2K —
Sleuth povvo/sleuth A 6-phase, 56-task OSINT workflow with a Python task runner, source grading, hypothesis analysis, templates and about 15 scripts.
With caveats Broad access 4/5 21 —
Firecrawl Skills firecrawl/skills Firecrawl's official catalog of 37 skills for web search, scraping, crawling and browser interaction, plus API integration and workflow recipes.
With caveats Check first 3/5 140 247K
Claude Patent Creator RobThePCGuy/Claude-Patent-Creator A Claude Code plugin and MCP server for US and EPO patent work: MPEP rule search, prior art search, claim checks, drafting and diagrams.
With caveats Broad access 3/5 206 —
Firecrawl CLI firecrawl/cli Firecrawl's command-line client and 13 agent skills for web search, scraping, crawling, interacting with pages and parsing files through the paid Firecrawl API.
With caveats Check first 2/5 674 660K
Best Agent Skills (Linkly) LinklyAI/best-skills A one-skill pack plus daily-updated CSV rankings of the top 100 agent skills, merged from skills.sh, ClawHub, Tencent SkillHub and social mentions.
With caveats Check first 2/5 676 —
Edgar Change Interpreter cmdrvl/edgar-change-interpreter A short skill that makes Claude compare SEC 10-K, 10-Q and 8-K filings for material changes and silent risks, with quoted evidence and fixed output sections.
With caveats Check first 2/5 7 —
research skill in mattpocock/skills Delegates a question to a background agent that reads primary sources and saves cited findings as a Markdown file in the repo.
With caveats No issues 1/5 306K 731K
Context7 upstash/context7 Up-to-date library documentation for coding agents through a hosted MCP server and ctx7 CLI, with short skills that tell the agent when to look docs up.
With caveats Check first 1/5 64K —
paper-context-resolver skill in lllllllama/RigorPilot-Skills A narrow helper skill that looks up one reproduction-critical detail in a paper, such as a dataset split, when the repo README leaves a gap.
With caveats Check first 1/5 486 480K

A Claude research skill tells an agent how to look things up and write them down: which sources to trust, how to cite them and where to save the result. Some read the open web, some query scholarly databases, and a few drive a hosted scraping service. We reviewed 22: 5 get an install verdict and 17 come with caveats.

The caveats follow from the work. Research means fetching outside content, and outside content is untrusted. The Claude Code MCP docs warn that servers that fetch external content can expose you to prompt injection, and the same applies to any skill that reads pages for you.

What to look for in a Claude research skill

First, where do the sources come from? Skills tied to public scholarly APIs such as OpenAlex, arXiv and Europe PMC are easier to reason about than skills that scrape social platforms. Second, does it verify? The better skills check citations against the source and save the evidence, not just a summary.

Third, count the keys and accounts. Several skills here depend on paid services or ask you to store provider keys. Fourth, check the license. One academic pack is non-commercial, which matters if you work for a company.

Finally, read the status. Of the 22, 3 are clean, 13 are marked “review” and 6 have broad access. Our security page lists what each flag means. The official guidance on third-party skills is to read scripts and dependencies before you enable them.

  • Co-Researcher has 15 skills, but the literature review skill is the substance: 21 scripts query OpenAlex, arXiv, Europe PMC, Crossref and DataCite to check citations, backed by 577 eval files. Setup uses uv, installed by a script. Status: review. Fields outside those indexes get less out of it.
  • last30days searches Reddit, X, YouTube, Hacker News and more for the last 30 days and writes a cited brief. It is one of the most complete research skills, with tests and a doctor command, and we label it broad access: it pre-approves unrestricted Bash, and an optional setup path can read your browser cookies for x.com after a consent prompt.
  • teach is clean and turns a directory into a multi-session learning workspace with a mission, resource list, notes and HTML lessons. It has about 883,000 recorded installs.
  • research is clean and tiny: 1.3 KB that sends a background agent to read primary sources and save cited findings as Markdown. It is a sound standard, but it adds little beyond asking for the same thing, which is why it has a caveats verdict.
  • Academic Research Skills cover deep research, paper writing, peer review and screening in five very detailed skills (about 52 KB each). The README licenses them CC BY-NC 4.0, so they are non-commercial, and the size uses a lot of context.

For web access, Firecrawl CLI is capable but is a front end for a paid hosted service: it sends your URLs and queries to the Firecrawl API, stores its key on disk and sends login usage metadata unless you opt out. See also browser skills for tools that drive a browser, and science skills for domain databases.

The category page lists every reviewed skill, and how we review explains depth scores.