190 个
为 MCP agent 提供 policy-as-code：在工具调用执行前拒绝高风险操作，以可验证证据证明实际运行内容，并在内核层面强制出站控制（eBPF/LSM，Linux）。确定性、离线优先、声明有界。Policy-as-code for MCP agents: deny risky tool calls before they run, prove what ran with verifiable evidence, and enforce egress in the kernel (eBPF/LSM, Linux). Deterministic, offline-first, bounded claims.
开源 DNS 与邮件安全扫描器：一个 MCP 端点、57 项检查、零安装，基于 Cloudflare Workers。Open-source DNS & email security scanner. One MCP endpoint, 57 checks, zero install. Cloudflare Workers.
Known Agents 的官方 Node.js SDK：AI agent 分析（bot 流量）、LLM 引荐来源追踪、自动生成 robots.txt。The official Node.js SDK for Known Agents: AI agent analytics (bot traffic), LLM referral tracking, automatic robots.txt
AI 辩论，代码裁定，亏损留在账面上。一款可移植的投资决策工作流插件与可验证 harness，已在真实的港股+美股组合上验证。AI argues. Code settles. The losses stay on the page. A portable investment decision-workflow plugin and verifiable harness, proven on a real HK + US portfolio.
AI Agent 的开源安全方案：macOS 与 Linux 上的内核强制出站控制、仅由运维持有的密钥、抗篡改审计。一条命令即可保护 Claude Code、Cursor 或任何 MCP harness。链路中无供应商介入Open-source security for AI agents: kernel-enforced egress control on macOS and Linux, keys only the operator holds, tamper-evident audit. One command protects Claude Code, Cursor, or any MCP harness. No vendor in the path.
Autobots | 由 NVIDIA NIM 驱动的去中心化模型 agent 集群。通过专用 6 文件控制架构编排高精度、端到端可鉴权软件开发流程，具备自动化安全审计、无冗余任务路由与自主状态管理能力。Autobots | A decentralized model agentic swarm powered by NVIDIA NIM. Orchestrating high-precision, end-to-end authenticated software development through a specialized 6-file control architecture. Features automated security auditing, non-redundant task routing, and autonomous state management.
面向 MCP 的 Agent 评估标准——对输出质量打分、捕获安全失败、强制成本预算。The agent eval standard for MCP — score output quality, catch safety failures, enforce cost budgets
自主式 AI agent 框架——基于 LangGraph 的 brain、swarm 编排、Arabic-first 设计，并具备人在环（human-in-the-loop）安全机制。Autonomous AI agent framework — LangGraph brain, swarm orchestration, Arabic-first, with human-in-the-loop safety.
跨平台 Codex 安装套件，包含专用 agent、精选 skill、默认安全的 MCP 配置、预览优先的安装流程以及清晰的验证步骤。A cross-platform Codex setup kit with specialist agents, curated skills, safe MCP defaults, preview-first installation, and clear verification.
你的 agent 能写出变更——却无法告诉你还有哪些内容依赖于它正在修改的代码。本工具从你自己的代码构建产品模型，然后对每个任务打分：变更内容、影响范围、波及的用户旅程与敏感数据——让运算结果直观呈现在屏幕上。本地运行，零依赖，无遥测。包含 MCP。Your agent can write the change — it can't tell you what else depends on the code it's touching. This builds a model of your product from your own code, then scores every task: what it changes, what that reaches, which user journeys and sensitive data are in the blast radius — arithmetic on screen. Local, zero deps, no telemetry. MCP included.
🤖 利用 TRPO、PPO、DPO、GRPO、DAPO、GSPO 等先进算法提升强化学习的稳定性与效率，优化策略训练。🤖 Enhance reinforcement learning stability and efficiency with advanced algorithms like TRPO, PPO, DPO, GRPO, DAPO, and GSPO for optimized policy training.
面向 Bugzilla 的安全加固 MCP server，使用 Rust 编写。Security-guarded MCP server for Bugzilla, written in Rust
基于 DNS、证书透明日志与未认证身份发现的公开元数据域情报。本地 Python CLI、版本化 JSON 与 stdio MCP server，无需凭证或主动扫描。Public-metadata domain intelligence from DNS, certificate transparency, and unauthenticated identity discovery. Local Python CLI, versioned JSON, and stdio MCP server. No credentials or active scanning.
Local-first、确定性的验证层，用于 AI 声明、记忆写入和风险操作。Local-first, deterministic verification layer for AI claims, memory writes, and risky actions.
基于 Rust 的 SIP 与 RTP 抓包、分析与安全工具，单二进制，仅依赖 libpcap。SIP & RTP capture, analysis, and security tool. One binary, one dependency (libpcap), built in Rust.
🔬 基于 AI 与研究论文对话，通过高级语义搜索与 RAG（检索增强生成）技术提取洞见与摘要。🔬 Chat with research papers using AI, extracting insights and summaries through advanced semantic search and Retrieval-Augmented Generation techniques.
自治 AI 的形式化框架——Neural Parliament、Ulysses Contracts、Identity Layer。提供参考实现与 Lean 验证的证明，Apache-2.0 许可。Formal framework for self-governing AI — Neural Parliament, Ulysses Contracts, Identity Layer. Reference implementation, Lean-verified proofs, Apache-2.0.
面向 Model Context Protocol 服务器的 OWASP MCP Top 10 安全扫描器OWASP MCP Top 10 security scanner for Model Context Protocol servers
确定性的 MCP 安全架构。以 FrozenNamespace 作为 Model Context Protocol 工具验证的可信根（Root of Trust）。Deterministic MCP Security Architecture. FrozenNamespace as Root of Trust for Model Context Protocol tool verification
面向会议的对话智能：将转录文本转化为摘要、决策、行动项、风险与机会，支持多租户自托管。FIAP NEXT Challenge 2026。Conversation intelligence for meetings: turns transcripts into summaries, decisions, action items, risks and opportunities. Multi-tenant, self-hosted. FIAP NEXT Challenge 2026.
基于 Python（Anthropic SDK）的多 Agent 系统，为私人教练生成个性化训练与饮食计划——内置免费规则引擎 + 可选 LLM 层、显式状态编排器、Streamlit 仪表盘与确定性安全校验器。A multi-agent system in Python (Anthropic SDK) that generates personalized workout routines and meal plans for a personal trainer — free rule engine + optional LLM layer, explicit-state orchestrator, Streamlit dashboard, and deterministic safety validator.
面向 agent 间通信的私有点对点身份、信任与验证基础设施——agent 注册中心、已验证身份、防篡改消息历史。Private peer-to-peer identity, trust, and verification infrastructure for agent-to-agent communication — agent registry, verified identity, and tamper-proof message history
实验性 Python 库：基于 dataframe 导航至研究问题，并内置统计与因果安全闸门Experimental Python library for dataframe-to-research-question navigation with statistical and causal safety gates.
Tox21 多终点毒性预测：可复现的研究与推理制品（冻结模型、FastAPI 服务、已审计的安全机制）Tox21 multi-endpoint toxicity prediction: a reproducible research and inference artifact (frozen model, FastAPI service, audited security)
1,716 道 AI/ML 面试题（含答题框架）、69 张架构图、按角色的学习路径，以及语音驱动的模拟面试工具。覆盖 LLM、RAG、agent、MCP/A2A、系统设计、MLOps、安全与云部署。1,716 AI/ML interview questions with answer frameworks, 69 architecture diagrams, role-based study paths, and a voice-enabled mock interview simulator. Covers LLMs, RAG, agents, MCP/A2A, system design, MLOps, safety and cloud deployment.
使 AI agent 能够在任何市场上自主创建、评估与进化 Skill，无需用户干预Enable AI agents to autonomously create, evaluate, and evolve skills across any marketplace without user intervention.
Hyper Terminal 2026：下一代 Web 技术 CLI ⚡ | 免费开源Hyper Terminal 2026: Next-Gen Web Tech CLI ⚡ | Free & Open Source
AI Agent 的认知操作系统——浏览器与计算机智能、持久化任务、记忆、恢复、证据与受治理的真实世界行动。A cognitive operating system for AI agents — browser & computer intelligence, persistent missions, memory, recovery, evidence, and governed real-world action.
基于 LlamaIndex、Redis 与 PII 脱敏的 RAG 搜索引擎。RAG search engine using LlamaIndex, Redis, and PII masking.
智利国家机构网络安全合规扫描器，对标 21.663 号法令及 ANCI 通用指令。结合主动网络扫描与声明式问卷，生成 PDF 差距报告以及面向 CSIRT Chile 的 JSON 报告。Escáner de cumplimiento de ciberseguridad para organismos del Estado chileno, alineado a Ley 21.663; y las Instrucciones Generales de la ANCI. Combina escaneo activo de red con un cuestionario declarativo para producir un informe de brechas en PDF y un reporte JSON listo para CSIRT Chile.
针对 AI Agent 的确定性审批、拒绝、取消、超时、重放及副作用安全测试 —— 无需 API key。Deterministic approval, rejection, cancellation, timeout, replay, and side-effect safety tests for AI agents—no API keys required.
本地 AI Agent 工具治理网关：不防提示词注入发生，防注入得逞后的高危工具调用后果（policy / approval / audit / MCP 双向治理）
围绕 LLM 人类价值观与多元对齐的精选论文、基准、数据集与工具合集。A curated collection of papers, benchmarks, datasets, and tools on human values in LLMs and pluralistic alignment.
面向 AI 生成可执行方案的 verifier-first 运行时，支持独立验证、对比与检索。A verifier-first runtime for independently verifying, comparing, and searching AI-generated executable solutions.
AI 安全测试——发现 LLM 应用、聊天机器人、AI Agent、MCP 服务器与 RAG 系统中的安全漏洞。对齐 OWASP（LLM、Agentic、MCP）与 NIST AI RMFAI security testing — find security vulnerabilities in LLM apps, chatbots, AI agents, MCP servers, and RAG systems. Mapped to OWASP (LLM, Agentic, MCP) and NIST AI RMF.
以安全为先、确定性合成的测试数据生成，基于 CSV 配置描述文件与白名单内的 Trino 元数据。Safety-first, deterministic synthetic test data generation from CSV profiles and allowlisted Trino metadata.