仓库/Skill 库

Perseus Vault 与 DeepSeek Harness 集成：MCP 配置示例 + 原生 pre-step memory-injection 插件Perseus Vault integration for DeepSeek Harness: MCP config example + native pre-step memory-injection plugin

面向 Agentic Dynamics 的实验工具：衡量 AI Agent 如何行为、恢复并产出已验证的结果Experimental instrument for Agentic Dynamics: measuring how AI agents behave, recover, and produce verified outcomes

本地、开源的推理框架，支持 Claude 和 Codex，包含 48 个编写好的系统，仅在需要更深层推理时选用。Local, open-source reasoning framework for Claude and Codex, with 48 authored systems selected only when deeper reasoning helps.

基于 LangGraph、LangChain、Groq 和 Tavily 构建的多步骤研究 Agent。分解复杂的研究问题，执行迭代式网络研究，评估证据，并生成结构化的最终报告。Multi-Step Research Agent built with LangGraph, LangChain, Groq, and Tavily. Decomposes complex research questions, performs iterative web research, evaluates evidence, and generates a structured final report.

项目名称：用于自动化文献综述与研究空白发现的自适应 AI 研究智能平台。Project Title: Adaptive AI Research Intelligence Platform for Automated Literature Review and Research Gap Discovery

有界研究 Agent：plan → retrieve → critique，具备步骤预算、执行轨迹、引用追溯与自由路径 UI，无需 API key。Bounded research agent: plan → retrieve → critique, step budgets, traces, citations, free-path UI. No API key.

在 Claude 中强制以引用优先的研究流程，通过结构化五阶段工作流验证事实、识别不确定性并消除幻觉。Enforce citation-first research with Claude to verify facts, identify uncertainty, and eliminate hallucinations through a structured five-phase workflow.

培育持久化数字生命的开放协议：协议、课程、研究问题、治理An open protocol for raising persistent digital beings: protocol, curriculum, research questions, governance

🔧 通过灵活的 prompt 模板与任务规范项目简化 AI Agent，实现高效的指令与记忆管理。🔧 Streamline AI agents with this flexible prompt template and task specification project for effective instruction and memory management.

Subconscious Memory 2026：基于混合语义搜索与 MCP 的 AI 编码上下文引擎。Subconscious Memory 2026: AI Coding Context Engine with Hybrid Semantic Search & MCP

插入到 MCP 服务器前方的安全与可信代理——策略拦截、rug-pull 检测、JIT 审批、哈希链审计。124 个测试。MCP 서버 앞에 끼어드는 보안·신뢰성 프록시 — 정책 차단, rug-pull 탐지, JIT 승인, 해시 체인 감사. 124 tests

一款 Specs AR 眼镜，可读取房间风水、进行评分、用 AI 重新渲染，并给出整改清单。基于 CLAD（Claude + Lens Studio MCP）构建。A Specs AR Lens that reads your room's feng shui, scores it, repaints it with AI, and gives you a checklist to fix it. Built with CLAD (Claude + Lens Studio MCP).

Python 包，通过 MCP、REST 与直接 Python 调用等方式，按 OHDSI 问题模板结构化研究问题。A Python package providing tools (MCP, REST, direct Python calls) for structuring research questions according to the OHDSI question templates.

Autonomous AI Algorithm Discovery Laboratory：一个有边界、可复现的研究平台，在固定算力预算下提出、实现、测试、拒绝、改进并报告 AI 系统的候选改进方案。Autonomous AI Algorithm Discovery Laboratory: a bounded, reproducible research platform that proposes, implements, tests, rejects, refines, and reports candidate improvements to AI systems under a fixed compute budget.

六个 Claude skills，覆盖写作、配图、润色、投稿与修改 Q1 期刊文章的全流程——其中一个可端到端运行 systematic review。适用任何领域、任何研究设计；采用拦截式 gate 而非堆砌式 checklist。Six Claude skills for writing, illustrating, polishing, submitting, and revising Q1 journal articles — including one that runs a systematic review end to end. Any field, any research design; gates that stop rather than checklists that sprawl.

📰 Spec-driven daily briefings on Claude Code cloud Routines — git as the only state, zero servers, zero API keys | 用 Claude Code Routines 每天自动生成主题简报，git 即全部状态，零服务器零密钥

可复现的研究流水线：将基于 Agent 的模型产出转化为有证据支撑的 LLM 高管报告。A reproducible research pipeline for converting agent-based model artifacts into evidence-backed LLM executive reports.

面向编码 Agent 的本地运行时，以合并质量为优化目标：由仓库自适应常驻模型生成补丁，大模型进行验证。单进程同时支持 Anthropic + OpenAI API，天生离线运行。A local coding-agent runtime that optimises for merge quality: a repo-adapted resident model generates patches, a large model verifies them. Anthropic + OpenAI APIs from one process, offline by construction.

向 Mac 上的 AI 编码 agent 展示 iOS 应用的实际运行状态——屏幕画面、应用内部状态、网络流量与日志，让检查 UI 变更不再依赖手动点击和截图循环。Shows what your iOS app is actually doing — the screen, the app's own state, its network traffic and its logs — to an AI coding agent on your Mac, so checking a UI change stops being a manual tap-and-screenshot loop.

关于 AI Agent 系统与编排的定向研究与文献综述的最终交付成果Final deliverables for directed studies and literature review on AI Agent Systems and Orchestration

将 OpenCode 转变为可复现的研究工具Turn OpenCode into a reproducible research instrument.

生产级 multi-agent 系统：接收研究问题、制定调查方案、从实时网络和私有文档语料中收集证据、对自身发现进行批判与事实核查，最终输出完整引用的 Markdown 报告。A production-grade multi-agent system that accepts a research question, plans an investigation, gathers evidence from the live web and a private document corpus, critiques and fact-checks its own findings, and delivers a fully cited Markdown report.

一个研究问题，一套跨 BV-BRC、BRC Analytics、PDN、NIAID Data Ecosystem、NCBI 资源及其他参与仓库的协同计划。One research question, one coordinated plan spanning BV-BRC, BRC Analytics, PDN, the NIAID Data Ecosystem, NCBI resources, and other participating repositories.

AI agent 辅助学术论文写作。An AI agent supports writing academic papers

基于 TypeScript 的独立框架，通过统一原语 Component 组合安全的个人 AI agent：manifest-first 发现、能力授予、正交模型路由与信息流控制。Standalone TypeScript framework for composing secure personal AI agents from one universal primitive — the Component. Manifest-first discovery, capability grants, orthogonal model routing, and information-flow control.

Yaroslav Vasylenko —— AI 系统、验证工程、可复现研究与 agentic workflows。Yaroslav Vasylenko — AI systems, verification engineering, reproducible research, and agentic workflows.

成本感知的 GitHub Copilot 工作流，用于清晰、有界且经过独立评审的软件变更。Cost-aware GitHub Copilot workflows for clarified, bounded and independently reviewed software changes.

Adobe Experience Manager 的 W3C WebMCP 集成 — 通过 navigator.modelContext 让浏览器 AI agent 可发现并调用 AEM Core Components。W3C WebMCP integration for Adobe Experience Manager - makes AEM Core Components discoverable and callable by browser AI agents via navigator.modelContext

lq 是面向 AI Agent 的独立 CLI 工具，用于解析、查询与修改 LyX 文档lq is a standalone CLI tool designed for AI agent to parse, query, and mutate LyX documents.

基于证据的论文检索与推荐 Agent，面向复杂研究问题。Evidence-grounded paper search and recommendation agent for complex research questions.

CELLO protocol 的官方客户端——提供 MCP server 及面向 OpenClaw、NanoClaw、IronClaw、ZeroClaw 等 agent 变体的原生适配器。The official client for the CELLO protocol — MCP server and native adapters for OpenClaw, NanoClaw, IronClaw, ZeroClaw, and other agent variants

基于多 Agent LLM 工作流的智能 AI 学术研究助手，可自动化完成学术论文发现、验证、文献分析、对比矩阵（含表格）、RAG 驱动的研究对话，并产出可直接用于稿件的文献综述An agentic AI academic research assistant that automates academic paper discovery, validation, literature analysis, comparison matrices with table, RAG-powered research chat, and manuscript-ready literature reviews using multi-agent LLM workflows.

基于 LangGraph、LangChain、Groq、ChromaDB 与 Streamlit 构建的多 Agent AI 研究助手，通过联网检索、获取 arXiv 论文、利用 RAG 索引 PDF 并生成有研究依据的回答，实现文献综述自动化Multi-agent AI Research Assistant built with LangGraph, LangChain, Groq, ChromaDB, and Streamlit. Automates literature review by searching the web, retrieving arXiv papers, indexing PDFs with RAG, and generating research-backed answers.

ResearchOS 是一款 MVP 级自主研究平台：输入研究问题，AI Agent 自动发现来源、验证证据、提取关键信息、交叉核对结论，并产出结构化的最终报告 — 全程实时可视化ResearchOS is an MVP autonomous research platform. Enter a research question; AI agents discover sources, verify evidence, extract key information, cross-check findings, and produce a structured final report — all visualized in real-time.

基于 Agno Agent 框架、Together AI 的 Qwen 模型与 Composio 工具构建的高级 AI 研究 Agent。该 Agent 通过生成研究问题、调用多个搜索引擎寻找答案，并结合 Google Docs 集成整理专业报告，帮助用户对任意主题开展全面研究An advanced AI research agent built using the Agno Agent framework, Together AI's Qwen model, and Composio tools. This agent helps users conduct comprehensive research on any topic by generating research questions, finding answers through multiple search engines, and compiling professional reports with Google Docs integration.

基于 n8n、Groq（Llama 模型）、Node.js、React 和 PostgreSQL 构建的多 Agent AI 研究流水线。用户提交研究问题后，由规划 Agent 拆解，由专业的研究/技术/成本 Agent 并行调查，由事实核查 Agent 验证，最终合成为带来源的结构化报告。A multi-agent AI research pipeline built with n8n, Groq (Llama models), Node.js, React, and PostgreSQL. A user submits a research question, which is broken down by a planning agent, investigated in parallel by specialized research/technical/cost agents, verified by a fact-checking agent, and synthesized into a structured report with sources.