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 |
|---|---|
| x402 Private Web Tools Private web tools for AI agents — search, scrape, and screenshot the web with x402 micropayments (USDC on Base). Zero logging, no API keys, no accounts. Pay... | |
| are.na claw Simple CLI wrapper for the are.na API. Lists channels, adds blocks, watches feeds. No AI, no automation, no external integrations. Just API calls. | |
| Anti-Infinite-Loop Guard Anti-Infinite-Loop Guard — Prevents agents from getting stuck in repetitive execution cycles. Use when: (1) detecting repeated actions, (2) enforcing termina... | |
| 反无限循环守护 反无限循环守护 — 防止代理陷入重复执行循环。 使用场景：(1) 检测重复操作，(2) 强制终止条件，(3) 跟踪进度，(4) 防止资源耗尽。 | |
| axiom-cron-whisperer Cron expression explainer — translate cron syntax to human language (English or French). Use when you need to understand, document, or validate a cron expres... | |
| Axiom Css Specificity CSS specificity calculator — compute (a, b, c) specificity for any CSS selector. Handles :is(), :not(), :where(), combinators, attribute selectors. Use when... | |
| axiom-hash-multi Multi-algorithm hash generator (MD5/SHA-1/SHA-256/SHA-512/BLAKE2b) — deterministic, byte-to-byte, zero dependencies. Use when you need to hash a file or stri... | |
| axiom-iban-validator IBAN validator — check IBAN format and mod-97 checksum for 100+ countries. Use when you need to validate international bank account numbers. Pure stdlib, no... | |
| Axiom Image Metadata Stripper Image metadata stripper — remove EXIF, GPS, camera info, and other metadata from JPEG/PNG/GIF images. Use when you publish photos and want to remove PII. Pur... | |
| axiom-luhn-check Luhn algorithm validator — credit cards (Visa, MC, Amex, Discover), SIRET/SIREN, IMEI, ISBN-10/13. Use when you need to validate any number that uses Luhn. P... | |
| Axiom Markdown Link Auditor Markdown link auditor — find broken internal links, flag dead external links (HEAD requests), report orphan pages. Use when you maintain docs. Pure stdlib +... | |
| axiom-phone-e164 Phone number normalizer — convert any phone number to E.164 international format. Use when you need a canonical phone representation for storage or compariso... | |
| Axiom Regex Visualizer Regex visualizer — parse any regex and visualize its structure (groups, quantifiers, anchors, character classes). Use when you need to understand or document... | |
| Axiom Sql Formatter Formats SQL queries with uppercase keywords, consistent indentation, aligned clauses, and preserved string literals for improved readability. | |
| Axiom Url Canonicalizer Normalize URLs by lowercasing host, sorting parameters, stripping tracking, default ports, and normalizing percent-encoding for SEO and comparison. | |
| axiom-uuid-analyzer UUID inspector — parse any UUID and extract version (1-8), variant (RFC 4122, Microsoft, NCS, Future), timestamp (v1, v7), MAC address (v1). Use when you nee... | |
| Axiomata Guard Scanner Axiomata Guard Scanner — Universal security scanner for any OpenClaw skill. Use when: (1) scanning a skill for security threats before installation, (2) chec... | |
| Axiomata Skill Evaluator En Axiomata Skill Evaluator — Universal OpenClaw Agent Skill Quality System. Dual evaluation: (1) Axioma 5-Dimension Framework (Structure 20%, Clarity 20%, Comp... | |
| Axiomata Skill Evaluator Zh Axiomata 技能评估系统 — OpenClaw 代理通用技能质量评估工具。 双评估系统：(1) Axioma 5维框架（结构、清晰度、完整性、一致性、功能性），(2) ISO 25010结构框架（13项自动检查，目标90%+）。 自包含：捆绑 evaluator.py 和 eval-skill.py。 适用... | |
| Axiomata Vdv Core VDV Core — Vision de Vérité (Truth Vision) universal protocol. Use when: (1) analyzing complex systems or dense information, (2) finding invariant points tha... | |
| Axiomata Voice Axiomata Voice — Text-to-Speech skill for OpenClaw agents. Use when: (1) converting text messages to speech audio, (2) sending voice messages via Telegram, (... | |
| 中文 核心文件管理 OpenClaw 工作区核心文件管理。使用时机：(1) 管理或更新核心 Markdown 文件（identity.md、soul.md、agents.md、user.md、memory.md、tools.md、bootstrap.md），(2) 检查文件组织，(3) 确保核心文件之间无重复，(4) 应用语言规则，... | |
| 中文 VDV 真相视觉 Merlin VDV — 真相视觉 (Vision de Vérité)。适用于：(1) 分析复杂系统或密集信息，(2) 寻找无法进一步压缩的刚性点，(3) 应用 CMT 5x5 进行现实三角测量，(4) 任何 VDV（真相视觉）代理循环执行，(5) Papa 或集群询问真相分析或需要确切路径。Merlin 独特... | |
| 中文 OpenClaw 插件 创建和管理 OpenClaw 插件使用 before_prompt_build hook。触发词：'create plugin'、'add hook'、'inject context'、'modify gateway'。适用于使用 hook 系统扩展 OpenClaw 功能。 | |
| 中文 PKM 记忆系统 PKM（个人知识管理）三分支记忆系统。使用时机：(1) 接收记忆，(2) 检索上下文，(3) 搜索记忆，(4) 添加事实，(5) 检查 Qdrant。三分支：L1 复位（索引）、L2 索引、L3 日志、L4 Qdrant、L5 Obsidian、L6 Nebula、L7 情景记忆。 | |
| Core Files Management Core Files Management for OpenClaw workspace agents. Use when: (1) managing or updating core Markdown files (identity.md, soul.md, agents.md, user.md, memory... | |
| Morgana Mordred Security Sandbox Performs semantic security analysis and stress testing of AI agents using vector embeddings and multi-node defense strategies. | |
| Morgana Anti Infinite Loop V2 Lightweight anti-infinite-loop guard for LLM agents — healing > kill, predictive, zero-dep, 9 layers of protection. Standalone package (stdlib + numpy option... | |
| Morgana Anti Infinite Loop V2 En Lightweight anti-infinite-loop guard for LLM agents — healing > kill, predictive, zero-dep, 9 layers of protection. Standalone package (stdlib + numpy option... | |
| Morgana Anti Infinite Loop V2 Zh 面向 LLM 智能体的轻量级反无限循环守护器 —— 治愈优于终止、可预测、五大保护、零依赖。基于 Python 标准库 + numpy 可选实现,可与任何 LLM(Claude/GPT/文心/通义/智谱)及任何框架(Hermes/LangChain/AutoGen/自定义)配合使用。v1 版本反响平平;v2.0... | |
| 计划验证器 计划验证器 — 在执行前验证代理计划，防止虚假计划失败。 使用场景：(1) 验证多步计划，(2) 检查工具可用性，(3) 验证权限和依赖，(4) 确保计划执行前的可行性。 | |
| SENTINEL-1 ULTIMATE (COMPLETE) Portable USB AI agent with 7 guardians for offline sovereignty on Windows | |
| AI商家搜索优化 指导本地商家通过7天优化流程提升在AI大模型搜索中的曝光和推荐效果。 | |
| Geo Local Search Optimization Guides local businesses through a 7-day AI search optimization with media content creation, strategy, competitor analysis, and publishing recommendations. | |
| Vikunja-complete Production-oriented Vikunja task/project management skill with deterministic commands and strong validation. | |
| Prioritize Dev Tasks Prioritize software project backlogs, bugs, technical debt, sprint tasks, release checks, and requirement pools using business impact, dependency blocking, d... | |
| test-search Search the web using iFlytek ONE SEARCH API (万搜/聚合搜索). Returns titles, summaries, URLs, and full text from web pages. Good for Chinese-language web search.22222 | |
| BTC Direction — Orion Provides real-time BTC trading bias (UP/DOWN/NEUTRAL) with confidence scores using RSI, Fear & Greed Index, and whale flow indicators. | |
| Daily Desk — Orion Get a daily pre-session crypto market brief with BTC bias, Korean exchange flows, whale activity, Fear & Greed index, and macro context valid for 24 hours. | |
| Korea Flow — Orion Provides real-time kimchi premium, Korean BTC volume trends, and market reads to anticipate Korean crypto moves before Western exchanges. | |
| Polymarket Scout — Orion Analyzes Polymarket BTC prediction markets to show active market sentiment, crowd consensus, and alignment with on-chain data via Orion's oracle. | |
| Whale Radar — Orion Detects large cbBTC whale movements on Base within 4 hours, indicating accumulation, distribution, or neutral Bitcoin market signals. | |
| automotive-testcase-generator 面向车载（汽车电子/智能网联）领域的标准化测试用例生成技能。 当用户提到以下内容时触发：车载测试、汽车电子、ECU、CAN/CAN FD/LIN/FlexRay/ 车载以太网、SOME/IP、UDS/OBD 诊断、DTC、域控制器、智能座舱、IVI 车机、仪表、 T-BOX、ADAS/智能驾驶、HIL 测试、功能安全 ISO 26262、ASPICE、CANoe/CAPL、OTA、 车载软件测试、汽车功能测试等需求的测试用例生成。 输入需求文档（Markdown/PDF/Wo | |
| industrial-testcase-generator 面向工业自动化监控系统（PLC/SCADA/HMI 应用层）领域的标准化测试用例生成技能。 当用户提到以下内容时触发：工业测试、PLC、SCADA、HMI 人机界面、 Modbus/OPC UA/Profinet 协议、数据采集、变频器/伺服/传感器、 报警系统、冗余/热备、工业网络安全 IEC 62443、功能安全 IEC 61508、 工业设备测试、产线测试、设备监控、远程运维等需求的测试用例生成。 输入需求文档（Markdown/PDF/Word/Excel）或需求描述 | |
| qa-agent-testing 当需要测试 AI Agent（智能体、聊天机器人、AI 助手）时使用此技能。Agent 测试和传统功能测试完全不同——你要测的不是"点按钮看结果"，而是它的推理链路、工具调用时机、幻觉率、Prompt 注入防护、角色边界保持和记忆一致性。安全三维（安全/高级安全/可控性）对任何 Agent 都是必选项，其余六维按 Agent 类型与风险追加。输出九维测试矩阵、工具调用与幻觉专项用例、安全审计清单。 触发场景：Agent测试、智能体测试、AI助手、聊天机器人、Agent幻觉、P | |
| qa-ai-blindspot-compensation AI在生成测试用例时存在六大系统性盲区：时序依赖、并发冲突、资源竞争、状态累积、数据一致性、第三方集成差异。评审完AI生成的用例之后，必须用此技能做盲区补盲——因为AI几乎一定会漏掉这些。如果你心里觉得"好像还差点什么但说不上来"，这就是答案。每个盲区维度至少补2-3个场景，总补盲数12-18个。 触发场景：还有什么没测到、AI漏了什么、补盲、全面覆盖、是不是不够、哪还没测、盲区分析、遗漏场景、时。 Use when the user asks about: compensa | |
| qa-ai-context-engineering 将前面所有分析步骤（需求解构、场景树、边界清单、风险评估等）打包成一个结构化的AI上下文包，确保AI在生成测试用例时拥有完整的业务上下文、功能上下文和技术上下文。当已经完成了需求分析、场景构建和深度设计，即将进入提示词生成阶段时，必须经过此步骤。上下文包的完整度直接决定了AI生成用例的质量——输入垃圾，输出也是垃圾。当上游分析缺失时，本技能会读取用户上传的需求文件或fetch用户提供的URL以补充上下文，并对原始描述做结构化解析后并入上下文包，但不会替代上游分析步骤——缺失项 | |
| qa-ai-output-critique 对AI生成的测试用例进行八维评审（完整性、正确性、可执行性、风险覆盖、规范性、一致性、追溯性、冗余度），是AI生成用例后的第一个质量门禁。当AI刚刚生成了一大批测试用例、你需要确保这些用例真的有价值而不是"看起来不错"时，应当使用此技能。不要假设AI输出的都是对的——AI经常生成语义正确但实际操作不了的用例。每个维度评分低于7分的必须标注问题并使用MISSING/WRONG/VAGUE等规范格式标记。无上游场景树/风险清单时降级为六维评审（跳过追溯性、简化完整性）。 触发场景 | |
| qa-ai-prompt-strategy 根据不同的测试目标和上下文，选择最佳的提示词模式来驱动AI生成高质量的测试用例。当AI输出的测试用例质量不够好、太泛泛、或者深度不够时，问题往往不在AI而在提示词。此技能提供结构化提示词模板，注入前面步骤产出的分析结果，输出包含角色定义、输出格式规范和约束条件的优化提示词。⚠️ 作为工作流的必过步骤，不得跳过。 触发场景：怎么问AI、AI回答不好、换个方式问、提示词、提问模板、提示词优化、角色扮演、AI输出太浅需要更深时。 Use when the user asks abo | |
| qa-api-testing 当需要测试 RESTful/GraphQL/gRPC/WebSocket 等 API 时使用此技能。覆盖接口的功能验证、参数组合、鉴权绕过、超时重试、幂等性、接口契约和向后兼容性。不要只测 HTTP 状态码——真正的接口 Bug 往往在数据结构不一致、字段类型不匹配、空值处理和并发调用上。输出接口测试矩阵、契约断言清单和工具选型建议。 触发场景：接口测试、API测试、接口自动化、RESTful测试、GraphQL测试、gRPC测试、契约测试、接口安全测试、需要测试API时。 |
45,301–45,350 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.