1,147 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 agents7,473Claude Code6,589Data and AI3,693Backend and APIs3,491Languages3,403Planning3,201Git and workflow3,131Code review2,731Memory and context2,354Testing2,335DevOps and cloud2,298Writing and docs2,284
Any agent Claude CodeCodexOpenCodeGitHub CopilotGemini CLICursorWindsurfKiro
| Project | Stars | Configures | Files | Always loaded |
|---|---|---|---|---|
| Airmail/airmail-mcp airmailapp.com Model Context Protocol (MCP) server for Airmail. Lets AI assistants manage emails, contact | 6 | Claude Code, Codex, OpenCode | 2 | 1,399 tok |
| RosMarinas/Research-Pi Research Pi is a multi-model Agent Runtime for computational research, organized around pr | 6 | Claude Code, Codex | 1 | — |
| jeffgreendesign/textrawl textrawl.com Turn your documents into agent memory. MCP server with hybrid semantic + keyword search ov | 5 | Claude Code, Codex, OpenCode | 4 | 3,324 tok |
| surendranb/find-research-papers-mcp MCP server for scientific grounding: search arXiv, OpenAlex, Crossref, Semantic Scholar, P | 5 | Codex, OpenCode | 1 | 600 tok |
| mblode/claude-code-search npmjs.com Search through your Claude Code history | 5 | Claude Code, Codex, OpenCode | 2 | 661 tok |
| xjli1972/book-brain A GBrain-shaped knowledge brain over open-source / free books. Markdown on disk, SQLite + | 5 | Codex, OpenCode | 1 | 1,375 tok |
| alsoleg89/doubt alsoleg89.github.io Turn contested questions into interactive, source-grounded evidence maps. | 5 | Codex, OpenCode | 1 | 257 tok |
| talesofai/okp cohub.live The context layer for AI agents — open, structured, searchable knowledge protocol (Domain | 5 | Codex, OpenCode | 1 | 1,386 tok |
| stefanoskarakasis/Product-Marketing-Skills stefanoskarakasis.substack.com PMM Skills Marketplace: 25+ agentic skills, commands, and plugins— from growth experiments | 5 | Claude Code, Codex, OpenCode | 2 | 1,719 tok |
| zebbern/pocmap zebbern.github.io MCP-native CVE, PoC & exploit discovery toolkit (Metasploit, Nuclei, GitHub & more) | 5 | Claude Code, Codex, OpenCode | 2 | 4,679 tok |
| tidyinfo/qmd A high-performance, private Markdown search engine with Hybrid Search (BM25 + Vector) and | 5 | Codex, OpenCode | 1 | 1,522 tok |
| tta-lab/organon Structure-aware tools for AI agents. Tree-sitter code editing, web page navigation, search | 5 | Codex, OpenCode | 1 | 1,892 tok |
| nolainjin/paper-verify pypi.org Standalone citation fact-checking tool for markdown/docs — extract, fetch, judge, score, r | 5 | Codex, OpenCode | 1 | 588 tok |
| bobaba76/Engram Local-first code intelligence and MCP-native review engine for Python and TypeScript repos | 5 | Codex, OpenCode | 1 | 736 tok |
| ChrystianSchutz/ThreadShelf Local-first archive, semantic search, and continuation for AI conversations across ChatGPT | 5 | Claude Code, Codex, OpenCode | 2 | 4,078 tok |
| brodewa369/roowet-semantic-vault-search Local RAG semantic search for Obsidian vaults — Ollama + LanceDB + MCP. Search your markdo | 5 | Codex, OpenCode | 1 | 1,221 tok |
| noblepayne/mcp-injector HTTP shim for injecting MCP tools into OpenAI-compatible chat completions | 5 | Codex, OpenCode | 1 | 7,776 tok |
| RobinOppenstam/sema-evals robinoppenstam.github.io Open, causal evaluations for content-addressed semantics and multi-agent coordination. | 5 | Codex, OpenCode | 1 | 286 tok |
| formin/spec-kit-harness Spec Kit extension: state-externalizing research harness (budgeted exploration, evidence c | 5 | Claude Code, Codex, OpenCode | 2 | 156 tok |
| retflo/retflo retflo.org Find out if you're right. Structured argumentation protocol for agents, researchers, educa | 5 | Codex, OpenCode | 1 | 3,723 tok |
| JessyTsui/scivly scivly.com Open-source AI research agent for tracking papers, generating summaries, and shipping dail | 5 | Codex, OpenCode | 1 | 618 tok |
| semoji-ai/auto_kairos | 5 | Claude Code, Codex, OpenCode | 20 | 4,892 tok |
| RamboLQX/LLM-Wiki-template | 5 | Claude Code, Codex, OpenCode | 3 | 8,845 tok |
| Steffen025/pag-extract Kongressprogramme als Datei statt 40 Browser-Tabs — liest das öffentliche Programm jedes m | 5 | Claude Code, Codex, GitHub Copilot, OpenCode | 4 | 1,767 tok |
| ccivlcid/VibeHarness | 5 | Claude Code, Codex, Gemini CLI, OpenCode | 24 | 3,278 tok |
| paperzilla-ai/paperzilla-skills Paperzilla Agent skills for OpenClaw, Claude Cowork/Code, Codex, ... | 5 | Codex, OpenCode | 1 | 33 tok |
| FelixSoderstrom/GIVERNY An agentic workflow I cooked up | 5 | Claude Code, Codex | 23 | 1,330 tok |
| kanazawazawa/copilot-agent-workspace | 5 | Claude Code, Codex, GitHub Copilot, OpenCode | 8 | 3,647 tok |
| johnmwangimegwe/kenya-climate-agent Reasoning Over Earth Engine: An AI Agent for Kenya's Climate Risk | 5 | Codex, OpenCode | 1 | 1,057 tok |
| lowtidebuild/ebook-writer A multi-agent system that transforms a topic into a professional, print-ready ebook. | 5 | Claude Code, Codex | 11 | 7,158 tok |
| bdhrs/vicaya Pāḷi and Dhamma research tool | 5 | Codex, OpenCode | 1 | 455 tok |
| trotsky1997/citeclaw Bun-migrated fork of Citoid with CLI and pipeline updates | 5 | Claude Code, Codex, OpenCode | 2 | 632 tok |
| lvzhaobo/ira Investment Research Assistant | 5 | Codex, OpenCode | 1 | 4,074 tok |
| Horosheff/Excalibur-2-Cloud | 5 | Claude Code, Codex, Cursor, OpenCode | 36 | 880 tok |
| ps-wiki/ps-wiki.github.io ps-wiki.github.io 📖 An open wiki for power systems engineers, researchers, and students | 5 | Claude Code, Codex, OpenCode | 3 | 2,014 tok |
| Gobindpreet9/tsx-stock-research Multi-stage parallelized investment trend research framework for TSX stocks | 5 | Claude Code, Codex, OpenCode | 3 | 3,128 tok |
| raphaelschwinger/torchrl-hydra-template Flexible and scalable template based on torch/rl + Hydra. Fostering rapid and reproducible | 5 | Claude Code, Codex, OpenCode | 2 | 12,500 tok |
| Mtrya/paper-scout Agent-native paper scouting workflow | 5 | Codex, OpenCode | 1 | 2,138 tok |
| TTAWDTT/elegant-researcher-skill A markdown-based skill bundle for end-to-end AI/ML research orchestration. | 5 | Claude Code, Codex, OpenCode | 2 | 1,916 tok |
| hgaddipati1118/slop-index The Slop Index: a benchmark ranking frontier LLMs by how much AI slop their writing contai | 5 | Claude Code, Codex, OpenCode | 2 | 1,981 tok |
| FoxsterDev/ai-research-hub foxsterdev.github.io Public AI operating layer for repo-attached engineering workflows, built around XUUnity. | 5 | Codex, OpenCode | 1 | 275 tok |
| valbert4/two-fold-transversal Repository for arXiv:2608.05688 | 5 | Claude Code, Codex, OpenCode | 2 | 1,620 tok |
| llsenyue/codex-history-vscode Manage your Codex history in VS Code. Features include session browsing, previewing, searc | 5 | Codex, OpenCode | 1 | 3,840 tok |
| Tongzh-SEU/ResearchScout | 5 | Codex, OpenCode | 1 | 3,528 tok |
| nsankar/Aletheia The Uncertainty Loop Agent | 5 | Claude Code, Codex, OpenCode | 3 | 4,330 tok |
| Nimble-Co/nimblenomicon nimblenomicon.pages.dev Nimblenomicon is an early WIP, content-first PWA concept for a smarter, searchable referen | 5 | Codex, OpenCode | 1 | 2,029 tok |
| asn-d6/morning_eprint Your sweet cup of eprint in the morning | 5 | Claude Code, Codex, OpenCode | 2 | 705 tok |
| Erica-byte-lee/codex-ai-interview-workflow-template A privacy-first Codex workflow for job analysis, interview research, and interview answer | 5 | Codex, OpenCode | 1 | 1,938 tok |
| raammurmu/runash.in-3 runash.in RunAsh is an AI Applied Research Lab focusing on live streaming innovation | 5 | Codex, OpenCode | 1 | 1,719 tok |
| fatecannotbealtered/jira-cli A CLI for AI agents to operate Jira Data Center — issues, sprints, boards, epics, projects | 5 | Codex, OpenCode | 1 | 252 tok |