仓库/Skill 库

FOREST 是一个面向长周期科学研究的 research agent，在此类任务中证据容易丢失、错误会累积、目标会发生漂移。它能够探索思路、阅读文献、形成假设、运行真实实验、分析结果、修正路径，并产出一篇完整论文——全过程置于一张可编辑的图中，支持审阅、分支与重定向，其中的论断与图表均与证据链接。FOREST is a research agent for long-horizon science, where evidence gets lost, errors compound, and goals drift. It explores ideas, reads literature, forms hypotheses, runs real experiments, analyzes results, revises paths, and builds a full paper—all in an editable graph you can inspect, branch, redirect, with claims and figures linked to evidence

开源、实时协作的 AI canvas——在无限画布上整合笔记、小应用与 agentOpen-source, real-time collaborative AI canvas - notes, mini-apps, and agents on one infinite board.

面向 macOS 的强约束 AI 编码 Agent 与开发环境自动化Opinionated AI coding agent and dev environment automation for macOS

Claude Code Agent Workflows Guide 2026：Skills vs Teams vs 成本分析Claude Code Agent Workflows Guide 2026: Skills vs Teams vs Cost Analysis

2026 年在社交信息流中通过实时聊天机器人模拟 AI 人格Simulate AI Personas in Social Feeds with Real-Time Chat Bots 2026

本地代码搜索，结合 BM25、向量相似度与 cross-encoder 重排序。使用 tree-sitter 解析 60+ 种语言，完全离线运行，返回包含文件路径、行号范围与符号元数据的结构化结果。使用 Rust 构建。Local code search combining BM25, vector similarity, and cross-encoder reranking. Parses 60+ languages with tree-sitter, runs entirely offline, and returns structured results with file paths, line ranges, and symbol metadata. Built in Rust.

Rove — 你的终端的 Agent 多路复用器。在隔离的 worktree 与持久化会话中并行运行编码 Agent 任务。Rove — the agent multiplexer for your terminal. Run coding agents on parallel tasks with isolated worktrees and persistent sessions.

面向 AI Agent 集群的终端 hypervisor：实时 pane 捕获、状态机模式检测，以及用于跨 WezTerm 协调 coding Agent 集群的 JSON API。Terminal hypervisor for AI agent swarms: real-time pane capture, state-machine pattern detection, and a JSON API for coordinating fleets of coding agents across WezTerm

开源、本地优先的 AI 长篇小说创作引擎：多智能体世界推演、按弧规划、RAG 长程记忆、逐章审核与断点恢复 | Open-source AI novel generator for long-form fiction.

一个 AI Agent，覆盖你的所有工作场景 —— 终端、桌面、移动端和聊天应用。Rust 内核。One AI agent, everywhere you work — terminal, desktop, mobile, and your chat apps. Rust core.

面向实验的 GenAI/RAG 优化器与工具包，基于 Oracle Database AI Vector Search 与 NL2SQLGenAI/RAG Optimizer and Toolkit for experimentation using Oracle Database AI Vector Search and NL2SQL

Windows 托盘上的 Kimi Code 用量监控：左键看额度，低额变红提醒。Tauri 2 + Rust，安装包仅 4 MB。

🦦 ADHDev — Agent 仪表板中枢。在统一仪表板中监控与控制 AI 编程 Agent，自托管开源。🦦 ADHDev — Agent Dashboard Hub. Monitor & control AI coding agents from a single dashboard. Self-hosted, open-source.

由 LLM 驱动的科研自动化 AgentLLM-powered agents for scientific research automation

可解释、本地优先的 AI Agent 记忆引擎。约 9 MB 的单一二进制在 VelesQL 下融合向量 + 图 + 列存；why() 返回每次召回背后的证据路径。无云依赖、无胶水代码 —— 支持服务器、浏览器、移动端与桌面端运行。The explainable, local-first memory engine for AI agents. One ~9 MB binary fuses vector + graph + columnar under VelesQL; why() returns the evidence path behind every recall. No cloud, no glue code — runs on server, browser, mobile and desktop.

为大型语言模型（LLM）和 RAG 系统转换并优化你的 markdown 文档，自动生成 llms.txt。Transform and optimize your markdown documentation for Large Language Models (LLMs) and RAG systems. Generate llms.txt automatically.

Claude Code、Codex、Grok 与 Cursor 的原生 Mac、iPhone 及 Apple Watch 配套应用。任务收件箱、手腕通知、agent 交接、配额小组件、可选语音与摘要。免费且开源。Native Mac, iPhone & Apple Watch companion for Claude Code, Codex, Grok and Cursor. Task inbox, wrist notifications, agent handoffs, quota widgets, optional voice and summaries. Free and open source.

面向 AI Agent 的任务范围授权。通过加密凭证约束工具与参数，在每次委托跳转中阻止权限提升，并生成关于允许或拒绝操作的有签名证据。Task-scoped authorization for AI agents. Cryptographic warrants constrain tools and arguments, prevent privilege escalation at every delegation hop, and produce signed evidence of what was allowed or denied.

2026 年实验室实验的顶级自主研究多 Agent 系统Top Autonomous Research Multi-Agent System for 2026 Lab Experiments

基于 Parquet 构建的嵌入式检索库。快速、高效且可扩展。Embedded retrieval library built on Parquet. Fast, efficient, and scalable.

100% 免费、开源的私人 Agent，适用于你的消息应用、工具和服务。100% free, open-source personal agents for your messaging apps, tools, and services

结合向量检索与时序知识图谱的 AI 记忆系统，内置面向 Agent 的认知引擎，支持记忆衰减、矛盾检测与 MCP 集成。AI memory system combining vector search with temporal knowledge graph. Built-in cognitive engine for agents. Supports memory decay, contradiction detection, and MCP integration.

将生产事故转化为结构化的 9 段式 LLM 响应（严重等级、根因、缓解措施、事后复盘）。附带 5 个场景的回归测试集 + LLM-as-judge 评估流水线。Turn a production incident into a structured 9-section LLM response (severity, root cause, mitigation, postmortem). Ships with a 5-scenario regression suite + LLM-as-judge eval pipeline.

Agent 记忆系统对比。Comparison of memory systems for agents

面向法律及文档密集型工作流的 AI 原生 IDE 工作区：文件、Agent、插件、内置支持修订追踪的文档编辑、OCR、证据链。律师的 VS Code。AI-native IDE workspace for legal and document-heavy workflows: files, agents, plugins, built-in document editing with tracked changes, OCR, evidence chains. VS Code for lawyers.

为 Coding Agent 提供快速、隔离的本地 React Native 和 Expo 环境。共享构建缓存、并行 worktree、独立设备以及安全清理。Fast, isolated, local React Native and Expo environments for coding agents. Shared build caches, parallel worktrees, owned devices, and safe cleanup.

面向 harness 工程、loop 工程、graph 工程，以及本地 AI Agent 工作流未来形态的桌面应用。Desktop app for harness engineering, loop engineering, graph engineering—and whatever comes next in local AI-agent workflows.

全天候验证的免费代理列表：HTTP、HTTPS、SOCKS4 和 SOCKS5 代理，每行标注国家、匿名性、延迟与在线率，每日多次更新。提供实时列表、代理检测、无密钥 API 与 MCP 服务器，访问 hproxy.com。Free proxy list, verified around the clock: HTTP, HTTPS, SOCKS4 and SOCKS5 proxies with country, anonymity, latency and uptime on every row, updated several times a day. Live list, proxy checker, keyless API and MCP server at hproxy.com.

小上下文、原生速度、全链路开源——一款开源终端 agent 运行时。单个 12 MB Go 二进制文件，零运行时依赖。Small context, native speed, open all the way down — an open-source terminal agent runtime. One 12 MB Go binary, zero runtime deps.

本地优先的桌面工作空间，用于持久化 AI 队友。支持 Codex、Claude、Grok 在专属工作空间中运行，具备任务队列、文件共享、浏览器控制与 Agent 间协作。A local-first desktop workspace for persistent AI teammates. Run Codex, Claude, and Grok with dedicated workspaces, task queues, file sharing, browser control, and agent-to-agent collaboration.

[EMNLP 2026] LightRSI：一个模块化运行时，用于在长程 LLM agents 中轻量部署递归自改进循环。[EMNLP 2026] LightRSI is a modular runtime for lightly deploying recursive self-improvement loops in long-horizon LLM agents.

Whip 是一个面向 Agent 的原生 SSH 移动客户端，从一开始便专为 Herdr 多路复用器设计。Whip is an agent-native SSH mobile client, designed specifically for the Herdr multiplexer from day one

动手学大模型全栈：CS336 中文精讲 · PyTorch 手搓 Transformer · 单卡复现 Pretrain/SFT/LoRA/DPO/GRPO/RLVR · PEFT/Agent/RAG 落地（含真实实验、曲线与复现脚本）

开源与闭源 AI agent 编排——一套完全可定制的 AI agent 队列，覆盖从头脑风暴到完全自主的项目维护各个层级AI agent orchestration for open and closed source — a fully customizable fleet of AI agents covering every level of project maintenance, from brainstorming to full autonomy

Vibe first. Then FastAgent——将本地 Agent 目录转化为实时服务：嵌入你的应用、部署到 GitHub、接入 Telegram 或任意渠道。Vibe first. Then FastAgent — turn a local agent directory into a live service: in your app, on GitHub, in Telegram, or any channel.

在你自有的服务器上运行一家由 Agent 组成"公司"。可启动具名 Agent（claude、codex、pi 等），将它们组织成组织架构图并共享待办列表，让它们交接工作，只在需要人类决策时推送通知到你的手机。MIT 协议。Run a company of AI agents on a server you own. Spin up named agents (claude, codex, pi…), put them on an org chart with a shared backlog, let them hand off work and ping your phone only when a human must decide. MIT.