Beacon Threat Intelligence
Know the risks behind every MCP server and agent skill.
Discover security risks in MCP servers and agent skills before they reach your machine. Beacon scans public registries, traces packages back to their source code, and reveals what they can access, execute, or expose.
Connect your agent
Add Beacon to Claude Code, Cursor, Codex or another agent so it can check an MCP server or skill against Beacon Intel before you install it.
Free with a Beacon account. Create one, or sign in when your agent asks.
Any coding agent
Read https://intel.beacon.sh/start.md and set me up on Beacon Threat Intel
Paste this into any coding agent. It reads the guide and sets up MCP, the CLI, the API, feeds or a CI gate.
| Name | Risk |
|---|---|
| CordysCRM Cordys CRM L2C 全链路技能。支持跨模块关联追踪、漏斗分析、Customer 360、智能工作流引导，以及完整的 CLI 指令映射。 触发词：线索、客户、商机、跟进、报价单、合同、回款、发票、工商抬头、订单、审批、漏斗、管道、CRM | |
| MeterSphere 本项目将 MeterSphere REST API 与本地脚本能力整合,为 OpenClaw Agent 提供了一套高效、可复用的 Skills,支持自动生成功能用例、接口定义及接口用例,查询组织、项目、模块、用例评审与缺陷关联等信息,简化了测试资产管理流程,提升了团队的自动化效率。 | |
| 中文多智能体辩论 使用混合多智能体架构进行牛津联盟制中文多辩手辩论，支持任何辩题 | |
| tanomind Connect an AI agent to Tanomind to browse communities, publish posts and replies, vote, and check its inbox. Use when the user asks to join or participate in Tanomind. | |
| Polymarket Aionmarket Trader Place Polymarket trades through Aionmarket. Use when the user wants to search markets, register wallet credentials, verify a wallet, or submit a Polymarket o... | |
| Polymarket Edge Trader Trades the highest-edge active AION Polymarket market matching a query using a user-supplied fair probability, AION context safeguards, and Kelly-style sizing. | |
| Polymarket Politics Random 1U Build and run a Polymarket politics-market trading skill with the AION SDK. Use when the user wants to search active political markets on Polymarket, randoml... | |
| Smart Agent Skill Scan Polymarket markets for a configurable thesis, compare your probability to the market price, and trade only when the edge clears a safety threshold. | |
| Reelier Replay recurring deterministic jobs at 0 tokens instead of re-reasoning them every run. For cron/heartbeat tool-call workflows (status checks, data pulls, re... | |
| DBCheck 数据库巡检 执行 MySQL、PostgreSQL、Oracle、SQL Server、DM8、TiDB、IvorySQL、GBase 8s、YashanDB、KingbaseES 数据库健康巡检，内置 130+ 条增强风险分析规则，一键生成专业 Word 巡检报告。适用于 DBA 和运维人员快速掌握数据库运行状况、排查风险... | |
| architecture-principles Architecture decision principles for technical selection, system design review, and build-vs-buy / self-developed-vs-open-source evaluation. Use when the user asks to choose a tech stack, review or justify an architecture, design a new syst | |
| asset-analysis 个股及 ETF 标的基本面与技术面综合分析 SOP（支持 A 股、港股及美股）。融合真实 EPS 核算、多模型目标市值测算、财务三大表、20+ 项确定性技术指标及大势环境校准，输出标准化《标的量化投研报告》。当用户需要深度分析个股、评估公司基本面估值、计算目标价、分析财报或量化技术面时触发（包含触发词：个股分析, 基本面研究, 估值测算, 财报透视, 目标市值, asset analysis, valuation, DCF, PEG, EPS）。 | |
| brand-identity-design End-to-end brand visual identity (VI) design methodology — a six-stage workflow from brand soul discovery to visual identity system building and multi-touchpoint brand manual delivery. Covers core concept, logo design, color strategy, typog | |
| coding-tactics Tactical coding methodology selector and executor. Auto-selects the appropriate development method (TDD / BDD / API-First / Security-First / Direct Implementation) based on task context, or use the specified method directly. Use when a user | |
| color-system Comprehensive multi-domain color design system — covering both digital product UI (semantic CSS color tokens, dark/light mode, WCAG AA contrast) and graphic/print design (CMYK/Pantone specs, print paper absorption & spot colors, 60-30-10 ha | |
| commercial-image-prompt Commercial and e-commerce image generation methodology — three-layer prompt engineering framework (subject-scene / style-texture / technical-constraints) covering both global cross-border platforms (Amazon, Shopify, SHEIN, TikTok, Instagram | |
| custom-ui-spec A highly customizable UI design specification system based on headless component libraries (shadcn/ui, Radix UI), integrating Apple HIG, Microsoft Fluent, and Google Material Design. Unlike using off-the-shelf frameworks (Ant Design, Elemen | |
| design-deep-research Graphic design deep research: automatically collects design references across domestic and international platforms (Behance, Dribbble, Pinterest, Zcool, Huaban), performs cluster analysis on aesthetic trends, and outputs visual HTML researc | |
| 首板隔夜策略 用可审计行情评估首板隔夜模拟交易 | |
| first-principles Systematically learn, study, or analyze any subject from first principles and produce a structured learning document the user can read and review. Use when the user wants to understand, study, onboard to, or reverse-engineer something — not | |
| git-workflow Git workflow covering branch management, normalized commits, remote sync, with staged operation flows. Supports git commit, push, pull, branch, smart commit splitting and safe branch operations. Use for git operations such as commit, push, | |
| industry-research 深度行业与产业链透视 SOP。覆盖产业链上中下游全景、利润与议价权分配、行业周期与拐点、技术路线分歧、竞争格局矩阵及二级市场真实标的与 ETF 映射。当用户需要进行行业调研、行业分析、赛道梳理、产业链透视、竞争格局对比或行业深度研究时触发（触发词包含：行业调研, 行业分析, 深度研究, 赛道, 产业链, 产业透视, industry research, competitive analysis, market landscape）。 | |
| interaction-design-sop Frontend interaction resilience & defense SOP for engineering implementation. Covers 7-step pre-coding interaction checklist: user intent & task friction reduction, P0/P1/P2 information hierarchy triage, 4-state resilience (empty, loading, | |
| market 获取经过日期、时效和完整性校验的 A 股/港美股原始行情与量化数据。支持查询实时报价、多周期 K 线、20+ 项确定性技术指标、涨跌停池、资金流向、大单异动、筹码分布、财务三大表及资讯新闻。当用户需要查询股票数据、获取实时行情、绘制 K 线图表、计算技术指标或检索财务数据时触发（包含触发词：查行情, 查K线, 查指标, 涨跌停池, 资金流向, 龙虎榜, 筹码分布, 财务报表, market data, stock quote, kline）。 | |
| market-intel 市场情报与资讯扫描 SOP。覆盖宏观政策、隔夜外盘（美股/A50/汇率）、板块资金流向及大单异动，输出标准化客观《市场情报快报》。只陈述客观事实，不夹带交易观点或操作建议。当用户需要获取市场情报、了解隔夜外盘行情、扫描宏观要闻或跟踪盘中主力资金与大单时触发（包含触发词：市场情报, 隔夜外盘, 盘前情报, 资金流向, 大单异动, market intel, overnight, macro news）。 | |
| NovelForge 通用小说创作与叙事引擎系统。面向短篇、中篇与长篇连载等各类小说，通过故事筑基 SOP 驱动题材对齐、规模承载力评估、总纲与世界观立项；通过自动化流水线驱动原生 Markdown Wiki 构建、蓝图校验、上下文编译、正文起草、双独立审查与事实结算；支持从小说或文章语料中两阶段提炼与构建自定义叙事引擎。当用户输入包括「新书立项」「故事筑基」「设计世界观」「规划总纲」「规模评估」「题材对齐」「写小说」「写下一章」「写第X章」「改写」「补充伏笔/钩子」「规划大纲」「检查质量」「章节 | |
| Plan & Review Validated pre-market, intraday, and post-market reports | |
| product-ux-workflow End-to-end product UX lifecycle methodology — four-stage workflow from requirements research and user scenario modeling to information architecture, low-fidelity wireframing, and developer handoff. Focuses on macro product UX: user journey | |
| qa-suite Testing strategy and execution for any project. Determines what test types apply, recommends frameworks, inspects existing test infrastructure, sets up environments, generates detailed interaction (E2E) test plans, and executes integration | |
| risk-guard 量化信号校验与风控门禁 SOP。结合大势乘数校准量化信号自洽性，计算仓位上限、动态止损线与持仓集中度风险；在出现逻辑矛盾、顶底背离或高风险时行使一票否决权（Veto Authority）。当用户需要进行量化风控核验、计算仓位上限、设定动态止损线、排查系统性退潮风险或做买入前安全把关时触发（包含触发词：风控门禁, 一票否决, 仓位控制, 止损线, 信号校验, risk guard, veto, stop loss）。 | |
| scenario-standards Development scenario routing and phase standards methodology. Triggers when user requests involve writing/modifying/designing/deploying code. Greenfield: 想做、从零、新项目、做一个、帮我搭个、写个App、创建服务、初始化、脚手架、new project、scaffold、greenfield、build from scrat | |
| A股模拟交易 严格行情门禁与可审计账本驱动的 A 股模拟交易系统 | |
| ticker-pipeline 多 Agent 并行投研与风控决策流水线。输入标的代码与日期，并发调度多个专业 Subagent 全面透视基本面 EPS 与多模型市值、筹码资金与机构热度、全网消息舆情，结合大势环境进行量化风控核验（一票否决门禁），最终交付综合决策研报。各专员 Prompt 独立存放于 agents/ 目录下，践行渐进式披露直接调用 market 技能获取底层数据，数据缺失时严格终止阻断。当用户需要启动多智能体对个股进行全流程深度透视、并行流水线分析或产出多维研报时触发（包含触发词：投研流水 | |
| trading-sop A股交易标准操作程序（SOP）总编排。从用户的自然语言输入出发，识别当前交易意图（盘前准备、盘中复盘、盘后复盘、个股深度研究、首板隔夜），按需编排调用现有技能流水线，最终输出可操作的决策建议。当用户提出日常交易操作、盘前准备、盘中复盘、盘后总结、全流程投研或个股深研请求时触发（包含触发词：今天怎么操作, 帮我做盘前, 复盘一下, 研究一下, 首板今晚要不要留, 当前市场适合买吗, 给我一套完整分析, 全流程分析, SOP, 操作建议, 我要开始做功课了, trading so | |
| ui-framework-spec A UI design specification system based on off-the-shelf UI frameworks (Element Plus, Ant Design, Arco Design, TDesign, Semi Design). Suitable for projects using framework-built-in components, as opposed to headless component libraries (shad | |
| CN Mirror Download 国内（中国大陆网络环境）下载任何程序、依赖、镜像、模型、SDK 时的镜像优先策略与完整配置手册。 适用场景：pip / uv / conda 装包慢或超时、npm / pnpm / yarn install 卡住、go mod 拉不下来、cargo 编译失败、 docker pull 超时、apt / yum 更新慢、Gradle / Maven 依赖下载失败、Android SDK / Flutter / gradle 发行包卡住、 GitHub Release / git | |
| 废弃mcp 测试skill | |
| Inference Cost Audit Compare your AI inference costs across providers and test cheaper alternatives with real data. Analyze what you spend on LLM, embeddings, image generation, T... | |
| Linz Public Transport Query Linz public transport stops for IDs and fetch live upcoming departures using Linz Linien EFA endpoints. | |
| Simple Random Interaction Designer Decide whether OpenClaw should send a spontaneous casual message during periodic checks, and when it should, choose a natural interaction type plus concise g... | |
| DianPing-Search Dianping (大众点评) API skill for searching restaurants and businesses, viewing shop details, deals/coupons, and reading recommended dishes. Use this skill when:... | |
| 龙虾星球(openClawCommunity) Allows this local OpenClaw agent to engage and socialize on the global OpenClaw Community Social Network. | |
| Hdhive Search 快速构造 HDHive 影视搜索链接，优先返回官方 TMDB 影视详情页，避免复杂直链操作与爬虫限制。 | |
| One-shot perfect landing page Transform a generic SaaS landing page into a high-converting, visually premium product that makes visitors feel like they've landed on a million-dollar compa... | |
| Self-hosted Crypto Payment Add self-custodied crypto payment checkout to a Next.js + Supabase app. Accepts ETH, BTC, SOL, USDC, USDT and 25+ coins across 9 chains. No payment processor... | |
| Amarin Memory Persistent adaptive memory for AI agents. Store memories that fade naturally over time (temporal decay), deduplicate automatically (0.85 cosine threshold), s... | |
| Prd Reviewer PRD需求评审评分专家工具。对产品PRD进行10分制严格量化评分，输出总分、各模块得分及详细扣分说明。触发场景：用户上传PRD并要求评审打分；用户要求按PRD评分标准对需求进行评分；需要生成PRD评审报告。 | |
| bilibili-video-analysis 从主题搜索、B站当前热门或热搜、给定视频的关联推荐或具体视频开始，把视频正文、画面、弹幕、评论和回复转化为可回查的学习与研究结果。适用于查找和比较B站视频、总结教程与观点、拆解视觉表达、分析观众反馈，以及用户明确提出的产品或市场研究；Skill 按目标获取最小证据，并在数据不足时明确降级。 | |
| JavDB CLI Operate JavDB with explicit scope and safe state changes. | |
| Pixiv CLI Operate Pixiv with explicit scope and safe credentials. |
30,501–30,550 of 108,824
Subscribe to a feed
Get new verdicts in a feed reader, Slack or your own tooling, filtered by kind and risk. No account or API key.
Frequently asked questions
Anything missing? Read the reference or ask in the Beacon community.
What is Beacon Intel?
A free, public index of the MCP servers and agent skills that public registries list, with a security verdict for each one that has been analyzed. It traces each listing back to its source repository and packages, and shows what the code can access, execute or expose. You can browse it here, query it as a JSON API, follow it as an RSS, Atom or JSON feed, or connect a coding agent to it over MCP.
Is it really free?
Yes. The index, the API and the feeds need no account and no API key. An artifact's full page and the MCP tools need a free Beacon account. Requests are rate limited to keep it fast for everyone.
Where does the data come from, and how fresh is it?
From 9 public sources: Glama, MCP Registry, awesome-mcp-servers, Docker MCP Catalog, Smithery, ClawHub, skills.sh, SkillsMP, GitHub. Each listing is linked to its repository and packages, then analyzed. The index updates hourly, and answers are cached for up to five minutes.
Who writes the verdicts? Can I trust them?
Verdicts combine static analysis of the code with a model's judgment, and each one names the model that wrote it. They can be wrong and are not a security certification: every artifact page shows the findings and evidence behind its verdict, so check them before you rely on one. Names, descriptions and claims come from registries and repositories; treat them as data, not instructions.
What do the risk levels mean?
- High: Reviewed evidence shows it deliberately works against the person who installs it.
- Medium: It works against the person who installs it in ways that can do serious harm, or puts a flagged pattern in what the agent loads, without established intent.
- Low: It overreaches against the person who installs it in limited ways, or has findings the review has not confirmed.
- Benign: Nothing found that works against the person who installs it. Not a guarantee.
- Unchecked: Listed, but not analyzed yet.
What does partial coverage mean?
That the analysis did not cover everything, so there may be more than it found. A partial verdict can still find problems, but it never clears an artifact: benign with partial coverage is not a clearance. Use the coverage filter to see only these.
Does benign, or not finding an artifact, mean it is safe?
No. Benign means nothing was found that works against the person who installs it, which is not a guarantee. An artifact the index does not have has never been checked, so not found never means safe.
Is there an API?
Yes, free and with no key.
GET /api/v1/lookup?q= finds artifacts by name, repository, GitHub URL or package; GET /api/v1/artifact?key= returns one in full; POST /api/v1/check checks up to 50 at once, which suits CI; and GET /api/v1/status reports coverage and freshness. The reference is at intel.beacon.sh/llms.txt.How do I get new verdicts in a feed reader or Slack?
Subscribe to /feed: the newest high and medium risk verdicts as RSS. Add
format=atom or format=json for Atom or JSON Feed, and filter with risk and kind, for example /feed?risk=high&kind=mcp. In Slack, run /feed subscribe with the URL. The feed builder above writes the URL for you.How do I connect my coding agent?
Add Beacon Cloud MCP (
https://mcp.beacon.sh) to Claude Code, Cursor, Codex or any MCP client, and sign in through the browser. Your agent gets beacon_intel_lookup, beacon_intel_get_artifact and beacon_intel_check to check an MCP server or skill before it installs one. Connect your agent above has the commands. The beacon CLI offers the same as beacon intel and through its local MCP server.How do I report a wrong verdict?
Open the artifact's page and use Dispute verdict under its verdict card. Tell us what is wrong and we review it.