agentsclimarketplace
Sign in

Shelves Research & analysis

Web research

808 rows from 349 repositories

Searching, reading and citing sources on the open web.

What Web research skills agree on

294 skill files read, by 292 of the 300 authors on this shelf whose files we hold, 2026-09-06

The middle one of the 36 measured here is ~1.6k tokens long, counted with cl100k_base

Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.

What they tell the agent to do

  1. Use web_search_exa for current information and broad discovery22 of 292 in 8 filesalso worded as use web_search_exa for broad queries; use web_search_exa for current info and news
  2. Cite every claim with a source21 of 292 in 18 filesalso worded as Cite every claim with a source; Cite a source for every claim
  3. Configure the Exa MCP server with an API key18 of 292 in 5 filesalso worded as Configure the Exa MCP server with an API key before searching; configure the Exa MCP server before use
  4. Use get_code_context_exa for code examples and API docs16 of 292 in 6 filesalso worded as Use get_code_context_exa for code examples and API usage; use get_code_context_exa for code examples and docs
  5. Verify exact tool names before depending on them13 of 292 in 4 filesalso worded as verify additional Exa tool names before depending on them; Verify exposed Exa tool names before relying on them
  6. Narrow results with site:, quoted phrase, and intitle: operators13 of 292 in 4 filesalso worded as Narrow results with site:, quotes, and intitle: operators; narrow results with site, quoted-phrase, and intitle operators
  7. Adjust tokensNum lower for snippets, higher for full context13 of 292 in 4 filesalso worded as Set tokensNum low for snippets, high for context; lower tokensNum for focused snippets, raise it for broad context
  8. Break the topic into 3-5 research sub-questions13 of 292also worded as Break the topic into 3-5 sub-questions; Break the topic into three to five sub-questions
  9. Get user confirmation after Phase 110 of 292 in 9 filesalso worded as Get user confirmation before Phase 2; Require user confirmation after Phase 1
  10. Prefer primary sources when available10 of 292also worded as Prefer primary sources over summaries; Prefer primary sources over secondary.
  11. Verify extracted metadata against original sources9 of 292 in 5 filesalso worded as Verify extracted metadata against the original publication; Spot-check extracted metadata against original sources
  12. Include an explicit limitations section in every report9 of 292 in 8 filesalso worded as Include an explicit limitations section; Include explicit limitations section in every report
  13. Extract every verifiable claim into a numbered list9 of 292 in 8 filesalso worded as extract every verifiable claim; extract every relevant claim into an explicit list
  14. Confirm current Exa docs and exposed tool surface before use11 of 292 in 2 filesalso worded as Confirm tool surface and Exa docs before relying
  15. Protect title capitalization with braces8 of 292 in 4 files

What they tell it not to do

  1. Never fabricate citations or invent sources19 of 292also worded as Never fabricate sources; Never fabricate citations
  2. Do not rely only on search snippets12 of 292 in 10 filesalso worded as Do not rely on a single search database; Don't rely only on search snippets
  3. Do not type BibTeX entries manually11 of 292 in 7 filesalso worded as Type BibTeX entries by hand; Never type BibTeX entries by hand
  4. Never make unsourced assertions10 of 292also worded as No unsourced assertions; Do not make unsourced assertions
  5. Do not accept extracted metadata without verification8 of 292 in 4 filesalso worded as Accept extracted metadata without verification; Do not accept extracted metadata blindly
  6. Do not submit without running final validation8 of 292 in 4 filesalso worded as Submit a bibliography without validation; Never submit without running validation
  7. Never give direct answers in Socratic mode7 of 292 in 6 filesalso worded as Direct answers in socratic mode
  8. Never run searches sequentially6 of 292 in 4 filesalso worded as Do not run searches sequentially; Don't run sequential searches in one context
  9. Never invent missing bibliographic fields6 of 292 in 5 filesalso worded as Do not invent missing bibliographic fields; Never invent missing citation fields
  10. Never fabricate references by mixing real papers5 of 292also worded as Never fabricate references from mixed real papers; Fabricating references by mixing real papers

What they expect to be installed

  1. WebSearch58 of 292 in 55 files
  2. WebFetch45 of 292
  3. web_search_exa24 of 292 in 10 files
  4. get_code_context_exa21 of 292 in 7 files
  5. python319 of 292
  6. CrossRef API18 of 292 in 14 files
  7. npx18 of 292 in 5 files
  8. curl17 of 292
  9. exa-mcp-server17 of 292 in 4 files
  10. AskUserQuestion16 of 292

What they ask it to produce

  1. Markdown research report file22 of 292 in 20 filesalso worded as Structured markdown research report; Markdown research report
  2. Research plan summary16 of 292 in 15 filesalso worded as research plan file; Research plan
  3. Research brief12 of 292 in 11 filesalso worded as Research Question Brief; brief.md research brief
  4. Structured markdown research report with citations12 of 292 in 11 filesalso worded as Structured markdown research brief with citations; Structured research report with inline citations
  5. JSON search results file11 of 292 in 7 filesalso worded as Search results JSON; JSON results file
  6. Citation validation report10 of 292 in 6 filesalso worded as Validation report; Citation verification report
  7. Verification report9 of 292 in 8 filesalso worded as Source verification report
  8. BibTeX bibliography file9 of 292 in 5 filesalso worded as BibTeX citation file; BibTeX bibliography files

When Web research authors say to reach for one

The situations these authors wrote into their own files, counted out of the same 292 authors, with the skills that name each one

How Web research skills are built

732 skill directories by 310 authors, read from their repositories’ own file trees 2026-08-05

The middle bundle among those shipping files is 4 files, 23.9 KB beside SKILL.md

Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.

The shape

  1. SKILL.md is the whole skill107 of 310 authors, 330 of 732 skills
  2. files ship beside it203 of 310 authors, 402 of 732 skills
  3. executable scripts ship inside110 of 310 authors, 203 of 732 skills

The folders they converge on

  1. references/116 of 310 authors, 193 of 732 skills
  2. scripts/97 of 310 authors, 189 of 732 skills
  3. agents/36 of 310 authors, 40 of 732 skills
  4. assets/33 of 310 authors, 46 of 732 skills
  5. examples/23 of 310 authors, 31 of 732 skills
  6. evals/22 of 310 authors, 36 of 732 skills

  • Light citation

    Light0305/Light-skills/skills/light-citation Skill

    521★ repo

    An AI workflow skill pack for research, competitions, and innovation projects.

  • Browser search

    Johell1NS/browser-search Skill

    486★

    A skill for AI agents: search the web with SearXNG, browse with Camofox, bypass protections with CloakBrowser. Anti-hallucination by design. Self-hosted, free, unlimited.

  • Blockrun mcp

    BlockRunAI/blockrun-mcp/io.github.BlockRunAI/blockrun-mcp MCP server

    426★ repo

    Web search, deep research, prediction markets & crypto data for AI agents. Pay per call via x402.

  • Rival search mcp

    damionrashford/RivalSearchMCP/plugins/rival-search-mcp-skills/skills/rival-search-mcp Skill

    119★ repo

    Deterministic research MCP server on FastMCP 3 — 5-engine web search, 9-platform social search, 6 academic DBs, news aggregation, entity profiles, conflict detection, document analysis. No API keys. No in-server LLM. Structured outputs for agent chaining.

  • Claude code skill

    beevibe-ai/beevibe-cto/examples/claude-code-skill Skill

    105★ repo

    Architecture Deep Research: deep research for strategic system design decisions.

  • Academic citation

    joshua-zyy/academic-paper-writer/skills/academic-citation Skill

    104★ repo

    面向 CS / AI / ML 领域的证据驱动、分节推进的论文写作 Agent Skill。

  • Deep research

    proyecto26/sherlock-ai-plugin/skills/deep-research Skill

    35★ repo

    Sherlock AI plugin; Research & Implementation Supercharged with a curated list of AI skills

  • Searchpin — Free Web Search for AI Agents

    telly6/searchpin/io.github.telly6/searchpin MCP server

    no license30★ repo

    Free web search for AI agents — multi-engine parallel search, smart ranking, no sign-up, no limits

  • Agri deep research

    PangenomeAI/academic-skills-food-nutrition/agri-deep-research Skill

    24★ repo

    Open, MIT-licensed food & nutrition science research skills for Claude Code, Codex, and MiniMax Agent — multi-agent literature/systematic review, journal-aware writing, peer review, and figures, plus author-guideline skills for 150+ journals. Initiated by the Food Science Group, University of Melbourne.

  • Scholar deep research

    Agents365-ai/scholar-deep-research/skills/scholar-deep-research Skill

    22★ repo

    8-phase literature-review pipeline. 7 federated sources, dedup, ranked retrieval, citation chasing, self-critique, 5 report archetypes.

  • Agentic search

    appautomaton/webmaton/skills/agentic-search Skill

    18★ repo

    Portable web-research and browser-automation SKILLs for Claude Code, Codex, and OpenCode — Playwright, Chrome DevTools, nodriver, and HTML-to-Markdown.

  • Perplexity web research

    xpepper/perplexity-agent-skill/perplexity-web-research Skill

    18★ repo

    A agent skill that leverages the Perplexity CLI for web research, deep reasoning, and independent validation.

  • Citation management

    TobiasBlask/open-paper-machine/scientific-skills/citation-management Skill

    16★ repo

    An autonomous LLM research agent that executes the full academic paper pipeline — from literature search to compiled PDF

  • PapersFlow

    papersflow-ai/papersflow-mcp/io.github.papersflow-ai/papersflow MCP server

    14★ repo

    Academic research MCP server for paper search, citation checks, graphs, and deep research.

  • Paper search with citation

    LivXue/paper-search-with-citation/paper_search_with_citation Skill

    no license11★ repo

    Paper search with citation skill for LLM agent

  • Citation formatter

    Marazii/research-co-pilot/skills/citation-formatter Skill

    9★ repo

    Comprehensive research co-pilot for Claude Code and claude.ai — 14 skills covering literature review, methodology, ethics review, data analysis, qualitative coding, brainstorming, manuscript drafting, replication design, grant writing, talk building, citations, survey design, and peer review. A peer collaborator, not a subservient assistant.

  • Academic citation

    XYD-GIS/paper-toolkits/skills/academic-citation Skill

    9★ repo

    中文学术写作工具集 · 38 个 prompt + 4 个 Claude Code skill

  • Citation Intelligence

    AutomateLab-tech/citation-intelligence/io.github.AutomateLab-tech/citation-intelligence MCP server

    8★ repo

    Check what Perplexity, Claude, ChatGPT, Gemini, and Google AI Overviews cite for any query.

  • Grok geo

    xuboboo/grok-geo/skill/grok-geo Skill

    5★ repo

    AI 搜索与生成式引擎优化（GEO）的品牌诊断 Agent Skill，内置 17 个主流 AI 搜索与问答引擎适配器，覆盖 ChatGPT、豆包、DeepSeek、通义千问（Qwen）、智谱 GLM / Z.ai、Kimi、文心一言、Claude、Gemini等平台，支持跨平台开展 GEO 查询与品牌表现分析。 系统基于 60+ 项可复现的确定性指标和 92 条合规规则，综合评估品牌在生成式搜索中的可见度、提及与引用表现、竞品差距、情感倾向及实体知识图谱质量，并输出可量化、可追踪、可执行的 GEO 优化建议，帮助品牌发现影响 AI 收录、引用与推荐的关键问题。

  • Deep research

    mayrsascha/deep-research-skill/skills/deep-research Skill

    5★ repo

    Agent skill that generates structured, context-aware Deep Research prompts and copies them to clipboard. Works with Gemini, Perplexity, ChatGPT, and any deep research tool. Compatible with Claude Code, Cursor, GitHub Copilot, and 18+ other agents.

  • Web search

    hec-ovi/websearch-skill/skills/web-search Skill

    5★ repo

    Keyless, uv-native web search + read for AI agents: ddgs multi-engine search with de-correlated rank fusion, Trafilatura extraction to paginated Markdown, plus keyless arxiv and github search. Self-hosted SearXNG with no Docker, SOCKS5 egress proxy, one-call init, and a doctor that self-tests every engine.

  • Citation verifier

    papersflow-ai/papersflow-skills/skills/citation-verifier Skill

    5★ repo

    Reusable PapersFlow research skills for literature search, citation graph exploration, and DeepScan monitoring powered by papersflow-mcp.

  • Deep research

    kongyo2/deep-research-skill/deep-research Skill

    no license4★ repo

  • Ieee citation

    TenWalk/ieee-skills/ieee-skills-claude/skills/ieee-citation Skill

    4★ repo

    Claude and Codex adapted IEEE communications manuscript skills for JSAC, TWC, TCOM, WCL, and CL.

  • Searxng websearch

    Mr-DS-ML-85/searxng-websearch Skill

    4★

    🔍 Zero-API-key web search for AI agents via self-hosted SearXNG. Works natively with Claude Code, OpenClaude, Gemini CLI, Codex CLI, Hermes & more. One SKILL.md, eight tools.

  • Deep research

    lyx2022518/sherlock-ai-plugin/skills/deep-research Skill

    no license3★ repo

    🛠️ Transform research papers into executable code with Sherlock AI Plugin, your essential tool for efficient literature review and implementation.

  • Aidress skill find verified web research agent

    Aidress-ai/aidress-skill-find-verified-web-research-agent Skill

    2★

    Find a Verified Web Research Agent. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for research agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).

  • Search1api

    superagents-lab/search1api-cli/skills/search1api Skill

    2★ repo

    Search1API CLI for web search, news, crawling, sitemaps, and trends—with OAuth 2.1 and an installable Agent Skill.

  • Academic researcher

    rakibulism/agent-skills-os/skills/academic-researcher Skill

    2★ repo

    THE UNIVERSAL AGENT SKILLS LIBRARY

  • Web search

    keenableai/keenable-mcp/ai.keenable/web-search MCP server

    2★ repo

    Live web search and clean-markdown page fetch over the Keenable web index.

  • SourceWeave Web Search

    MRNAQA/sourceweave-web-search/io.github.MRNAQA/sourceweave-web-search MCP server

    2★ repo

    Fully local MCP server and CLI for web research

  • Duckduckscrape

    rafaelpierre/duckduckscrape/.claude/skills/duckduckscrape Skill

    no license1★ repo

    A lightweight CLI that provides agents with web search powers via DuckDuckGo.

  • Deep research revision

    foundry-works/foundry-research/skills/deep-research-revision Skill

    1★ repo

    Claude Code agentic researcher designed for academic research, supporting 20+ academic and web source providers

  • Deep research

    B143KC47/deep-research-skill/deep-research Skill

    1★ repo

    Evidence-first deep research skills for AI agents, with source tracking, citations, contradiction checks, and uncertainty-aware synthesis.

  • Web research cascade

    belschak/web-research-cascade Skill

    1★

    Claude Code skill that escalates blocked web fetches (WebFetch -> Jina Reader -> Firecrawl -> your browser) so primary sources get read instead of silently skipped.

  • Medical fact check

    kgraph57/evidentia/skills/medical-fact-check Skill

    1★ repo

    Open-source medical citation verifier: catches fabricated DOI/PMID/NCT references with CLI, MCP, and Claude Code skill.

  • Deep research

    hashbulla/deep-research Skill

    1★

    A human-gated, 7-phase agentic deep-research skill for Claude Code. Fans out across the open web (Tavily), GitHub, academic open-graphs (OpenAlex/arXiv/Semantic Scholar), and live library docs — then grades every source on the NATO Admiralty A–F scale, runs deterministic quality gates, and emits four cited artifacts.

  • Deep research

    sergeyizmailov/Claude-Skills/skills/deep-research Skill

    1★ repo

    Open Agent Skills for frontend engineering, secure coding, research, automation, and Meta Ads workflows.

  • Deep research

    FMATheNomad/deep-research-skill/skills/deep-research Skill

    1★ repo

    Massive autonomous web research for AI coding agents. ⭐ Free Firecrawl alternative. No API keys. Open source. Sponsor ❤️

  • Citation faithfulness

    Hash-7777/Hash-Medical-Reasearch-Agent-Skills/skills/citation-faithfulness Skill

    1★ repo

    Drop-in agent skills that make an AI appraise medical literature like a reviewer — grade every citation against its source, run reproducible PRISMA searches, pool studies safely, and defend RAG against prompt injection. Framework-agnostic, MIT.

  • Searxng search

    FactrueSolin/searxng-skills/skills/searxng-search Skill

    1★ repo

    A portable Agent Skill for searching the web through a user-configured SearXNG instance, with source-aware research workflows and a companion Python CLI.

  • Web search

    PROJECT-B-26/groundroute-mcp/ai.groundroute/web-search MCP server

    1★ repo

    Web search for AI agents — one tool across 6 engines, routed to the cheapest + cached.

  • Citation Verification

    guptaprakhariitr/verification-mcp/io.github.guptaprakhariitr/verification-mcp MCP server

    no license1★ repo

    Real-time fact-check, citation verification, and source-freshness for AI agents.

  • Deep research skill

    Chris1220-cmd/deep-research-skill Skill

    0★

    Free Perplexity/Gemini-level deep research for Claude Code — 20-40+ sources, domain boost, source credibility, then generate podcasts, videos, slides, quizzes from findings

  • Perplexity research

    rainearcher/quiver/skills/perplexity-research Skill

    0★ repo

    Claude Code plugins for deep GTM research, competitive analysis, and more — by Raine Soriano

  • Grounded research

    moonpiesheldon1337/grounded-research Skill

    0★

    A Claude skill that enforces citation-first research discipline. Classifies claims, verifies before writing, signals confidence levels, and audits output for hallucinations. Works with Claude.ai, Claude Code, Cursor, Codex, and Gemini CLI.

  • Truth shield

    BAS-More/truth-shield Skill

    0★

    Stop Claude from lying to you. A Claude Code skill that fact-checks every claim against real sources — code files, live docs, web search — and flags anything it can't confirm.

  • Ai citation optimizer

    excelle/seo-geo-toolkit/craft/ai-citation-optimizer Skill

    0★ repo

  • Grounded research

    ahoyvarun/grounded-research/grounded-research Skill

    0★ repo

    A Claude Skill that verifies research writing - catches fabricted citations, unsupported claims and hidden contradictions

  • Aeo llm content planner

    search-atlas-group/amm-founding-circle/skills/aeo-llm-content-planner Skill

    0★ repo

    The AMM founding-circle home base: 36 Claude skills (AEO/SEO + agentic engineering + security), the agentic ladder, playbooks, and automations. No paid APIs required.

  • Citation formatter

    prvthmpcypher/skills-education/skills/citation-formatter Skill

    0★ repo

    14 Claude skills for learning and teaching. Flashcard generation, concept explanation, study plans, exam prep, and research summaries.

  • Skill

    Fizzokid1/truthful-research/skill Skill

    0★ repo

    Verification-first research skill for Claude: no quote, no citation. Plus the rubric to grade any research agent.

  • Sift

    indigokarasu/sift Skill

    0★

    Sift: research synthesis engine — retrieves, verifies, and synthesizes information from the web with source reliability scoring.

  • Citation survey skill

    ultimatile/citation-survey-skill Skill

    0★

    Agent skill to survey citing or referenced papers using the OpenAlex API.

  • Apa citation

    cmdecker95/skills/skills/apa-citation Skill

    no license0★ repo

    🦾 My own little curation of skills

  • Deep research

    ankitgoyalio/life-skills/deep-research Skill

    0★ repo

    A collection of Agent Skills that extend AI agents for health, fitness, nutrition, deep research, and practical life workflows.

  • Citation reference reviewer

    ULudo/codex-science-skills/skills/citation-reference-reviewer Skill

    no license0★ repo

    Codex skills for scientific manuscript drafting, peer-review assessment, technical rigor checks, citation auditing, and research writing workflows.

  • Crawlit skill

    arufian/crawlit-skill Skill

    0★

    Agent skill pack for Crawlit web scraping, crawling, and site mapping

  • Gemini web search

    guillochon/gemini-google-web-search-skill/skills/gemini-web-search Skill

    0★ repo

  • Ks deep claim audit

    KaiSky0823/ks-deep-claim-audit Skill

    no license0★

    Claude Code skill · 对别人宣称的 GitHub 项目/AI工具/产品/数据·性能·热度做深度地毯式核查,客观判真伪夸大,出逐条 verdict 报告

All 808 in the catalog