1,201 projects whose own agent configuration covers research. These are not mods to install — they are how the maintainers of these repositories tell an agent what to do, and what to avoid.
All projects Building agents8,442Claude Code6,875Data and AI3,584Git and workflow3,574Backend and APIs3,550Languages3,468Planning3,315Code review2,876Testing2,653DevOps and cloud2,462Writing and docs2,461Memory and context2,373
Any agent Claude CodeCodexOpenCodeGitHub CopilotGemini CLICursorWindsurfKiro
| Project | Stars | Configures | Files | Always loaded |
|---|---|---|---|---|
| davidmosiah/wellness-nourish wellness.delx.ai Local-first nutrition MCP for Claude/Cursor: USDA food search, barcode + photo, meal loggi | 11 | Codex, OpenCode | 1 | 173 tok |
| fmschulz/omics-skills fmschulz.github.io | 11 | Codex, OpenCode | 1 | 3,325 tok |
| apresmoi/jianglens jianglens.com Agentic Organization Research Project - Jiang Lens is an independent research and reading | 11 | Codex, OpenCode | 1 | 2,332 tok |
| PsychQuant/che-apple-mail-mcp Apple Mail MCP server — 53 tools with SQLite-powered millisecond search across 250K+ email | 10 | Claude Code, Codex, OpenCode | 2 | 1,980 tok |
| alexylon/sofos-code Terminal-based AI coding assistant with Claude/OpenAI support, secure local tools, web sea | 10 | Claude Code, Codex, OpenCode | 2 | 4,931 tok |
| zhangyu94/humanfile Declare boundaries for coding agents with .human files. | 10 | Claude Code, Codex, GitHub Copilot, OpenCode | 9 | 1,003 tok |
| davidchicano/presuntamente Archivo público e interactivo de casos de corrupción en España. Cronologías, actores, docu | 10 | Claude Code, Codex, OpenCode | 11 | 11,070 tok |
| dongzhuoyao/tao-research-skills | 10 | Claude Code, Codex, OpenCode | 2 | 2,484 tok |
| joelhooks/content-grill 🔥 Local context plane for topic grilling — slurp Brain/X, search fast, topics only (never | 10 | Claude Code, Codex, OpenCode | 2 | 1,101 tok |
| tjcoder-labs/cli tjcoder.com Terminal-first research harness exploring generative UIs and model-directed presentation a | 10 | Codex, OpenCode | 1 | 1,269 tok |
| Zotero-Cat/Zotero-Cat zoterocat.org Zotero-Cat is a Zotero item-pane assistant for reading, summarizing, reviewing, and discus | 10 | Codex, OpenCode | 1 | 7,980 tok |
| Rai220/money_alert_ai An AI bot runs the Telegram channel @money_alert_ai, assessing the financial stability of | 10 | Codex, OpenCode | 1 | 1,966 tok |
| Shiien/research-closure-harness | 10 | Claude Code, Codex, OpenCode | 4 | 1,468 tok |
| H-Ali13381/historical-bayesian-html-briefing Agent skill for historical analogies, Bayesian forecasting, and self-contained HTML briefi | 10 | Codex, OpenCode | 1 | 457 tok |
| rolznz/sovereign-brain sovereignbrain.me A local and private sovereign second brain for brainstorming, decision making, working on | 10 | Codex, OpenCode | 1 | 757 tok |
| catoncat/xsearch One-shot structured search CLI and Agent Skill for OpenAI-compatible Grok model endpoints | 10 | Codex, OpenCode | 1 | 270 tok |
| gwas-catalog/mcp ebi.ac.uk GWAS Catalog Model Context Protocol server, supporting natural language queries of associa | 10 | Codex, OpenCode | 1 | 366 tok |
| dinosn/security-knowledge-base Local-first, evidence-led security knowledge base with a model-neutral JSON CLI, immutable | 10 | Claude Code, Codex, Gemini CLI, OpenCode | 3 | 1,401 tok |
| DanielGouveiah/glasses daniskills.vercel.app Put your glasses on. Evidence-first planning and bounded implementation diagnosis for AI c | 10 | Claude Code, Codex, OpenCode | 2 | 73 tok |
| cosen1024/awesome-academic-skills A curated, machine-validated catalog of public Agent Skills for academic research workflow | 10 | Codex, OpenCode | 1 | 474 tok |
| cinicu/opencode-dotfiles OpenCode dotfiles — 19 AI agent skills + 8 MCP servers + persistent memory (memory-keeper) | 10 | Codex, OpenCode | 1 | 218 tok |
| quzhi1/HistoryAgentSkills Use Agent Skills to search history source and answer questions | 10 | Claude Code, Codex, OpenCode | 4 | 1,363 tok |
| whenpoem/aiscientist An AI-driven research workflow for generating ideas, running experiments, and writing scie | 10 | Claude Code, Codex, OpenCode | 6 | 1,865 tok |
| raunakkathuria/buildwright Approve intent, ship autonomously. Agent-first autonomous development workflow for Claude | 10 | Claude Code, Codex, OpenCode | 2 | 108 tok |
| Tx1207/hello-scholar In Agent-related practice, I found that blindly piling on a large number of skills does no | 10 | Claude Code, Codex, OpenCode | 2 | 4,020 tok |
| adsabs/scix-mcp A Model Context Protocol (MCP) server for the NASA Astrophysics Data System (SciX) API | 10 | Codex, OpenCode | 1 | 828 tok |
| AndekQR/fitatu-mcp-unofficial Unofficial Model Context Protocol server for Fitatu, with tools for food search and meal p | 10 | Codex, OpenCode | 1 | 2,556 tok |
| oghie/skillsets My personal collection of agent skills for software, security, product and research engine | 9 | Codex, Gemini CLI, OpenCode | 2 | 734 tok |
| lbj96347/compete A Claude code CLI plugin for discovering competitors and conducting comparisons to identif | 9 | Claude Code, Codex, OpenCode | 2 | 2,056 tok |
| huangzhonglv/hep-workflow A skill-based agent workflow for high-energy physics phenomenology: from model proposal th | 9 | Claude Code, Codex, OpenCode | 8 | 3,316 tok |
| pengzhangzhi/superpowers-ml Skills that make your AI agent do ML research like a skeptical scientist, not a hype machi | 9 | Claude Code, Codex, Gemini CLI, OpenCode | 3 | 1,913 tok |
| kasssandr/archilles RAG for researchers: page-level citations from your personal library, LLM access via MCP. | 9 | Claude Code, Codex, OpenCode | 2 | 3,480 tok |
| speakeasy-api/docs-mcp speakeasy.com Offline-first, performant MCP server for your docs | 9 | Claude Code, Codex, OpenCode | 2 | 3,134 tok |
| gaelic-ghost/socket macOS Agent Workflows, Research, & Experiments | 9 | Codex, OpenCode | 1 | 741 tok |
| hutzelmann/thesis-proposal-skills skills.sh Write a convincing thesis proposal with support from AI | 9 | Claude Code, Codex, OpenCode | 3 | 7,245 tok |
| pppop00/Equity-Research-Company Harnessed Equity Research Team for fundamental analysis reports. | 9 | Claude Code, Codex, OpenCode | 3 | 4,536 tok |
| ajhcs/healthcare-data-mcp 18 local MCP servers for public healthcare market intelligence: facilities, ownership, qua | 9 | Codex, OpenCode | 1 | 54 tok |
| ericxuzhesheng/Convertible-Bond-Pricing-Research This project builds an absolute pricing framework for Chinese convertible bonds using BS a | 9 | Codex, OpenCode | 1 | 4,780 tok |
| lzehrung/codegraph CLI, MCP and TS library that turns source code into a map of files, symbols, references, a | 9 | Claude Code, Codex, GitHub Copilot, OpenCode | 3 | 3,270 tok |
| VelvetSP/web-retrieval-mcp MCP server for AI agents: high-fidelity web search (Exa) + tiered web fetch (Exa→browser→F | 9 | Codex, OpenCode | 1 | 196 tok |
| Luminaire1337/mtasa-docs-mcp MCP server for MTA:SA documentation with intelligent search and caching | 8 | Codex, OpenCode | 1 | 5,603 tok |
| lililib/ai-session-search Local-first full-text search and reusable context library for AI coding sessions across Cl | 8 | Codex, OpenCode | 1 | 113 tok |
| jmpijll/fortimanager-code-mode-mcp MCP server for FortiManager using the Code Mode pattern — 2 tools (search + execute) with | 8 | Codex, OpenCode | 1 | 1,701 tok |
| nutdnuy/sec-opendata-th Python client + Claude Code plugin for the Thai SEC OpenAPI (api.sec.or.th): fund factshee | 8 | Claude Code, Codex, GitHub Copilot, Gemini CLI, OpenCode | 4 | 1,912 tok |
| flanliulf/aiforge Sync AI coding configurations from any Git repository to Copilot, Claude Code, Cursor, and | 8 | Claude Code, Codex, Gemini CLI, OpenCode | 53 | 1,374 tok |
| bezata/kObsidian Filesystem-first MCP server for Obsidian vaults with an LLM-Wiki layer on top. | 8 | Claude Code, Codex, OpenCode | 2 | 928 tok |
| chrisgagne/grounded-forge Project once, retrieve distillations, create tools: a pre-loading retrieval architecture a | 8 | Claude Code, Codex, OpenCode | 4 | 5,455 tok |
| counterposition/pi Skills, extensions, and packages for the Pi coding agent | 8 | Claude Code, Codex, OpenCode | 2 | 269 tok |
| viplee110/economics-paper-reviewer Portable AI pre-review skill for economics papers across Codex, Claude Code, Cursor, and o | 8 | Claude Code, Codex, OpenCode | 2 | 425 tok |
| alsoleg89/doubt alsoleg89.github.io Turn contested questions into interactive, source-grounded evidence maps. | 8 | Codex, OpenCode | 1 | 257 tok |