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 |
|---|---|
| 沙箱智能体对话 Chat Chat with the dlazy sandbox agent — a project-scoped assistant that runs skills end-to-end over multiple turns. Discover skills and projects with dlazy skills list / dlazy projects list. 与 dlazy 沙箱 agent 对话 —— 一个以项目为单位、可端到端运行技能的多轮助手。用 dlazy | |
| 大模型对话 Claude Opus 4.7 Anthropic's flagship multimodal model. Strong at long-context reasoning, code generation, and complex tool orchestration. Supports text, image, and video inp... | |
| 文本生成 Claude Sonnet 5 Anthropic's latest Sonnet — near-Opus quality on coding and long-horizon agentic work at Sonnet cost. Strong at reasoning, code generation, and complex tool orchestration. Supports text, image, and video inputs. | |
| 知识付费讲师数字人 Course Avatar Knowledge-course / online-teaching avatar lecturer — 16 landscape base-portrait recipes plus the slide-first pipeline that keeps a 20-minute lesson under 3000 credits. Use when the user wants a digital human to teach a paid course, a traini | |
| AI 内容检测 AI Detect Detect whether an image, video, or audio file is AI-generated — including visual deepfakes and the likely generator model (Midjourney, Stable Diffusion, Sora, etc.). Returns confidence scores you can threshold. 检测图片、视频或音频是否由 AI 生成：包含人脸 deep | |
| 文档转视频 Doc to Video doc to video, word to video, markdown to video, document to video — parse the document, outline, storyboard, voiceover, build, validate. Use when the user gives a Doc / Word / Markdown file and wants an explainer, report broadcast, or train | |
| 语音合成 豆包 Doubao TTS Synthesize text into natural and fluent speech using Doubao TTS. 使用豆包 (Doubao) TTS 文本转语音模型，将文字合成为自然流畅的语音播报。 | |
| 电商带货数字人 Ecommerce Avatar Ecommerce livestream avatar / AI spokesperson for product selling videos — 18 production-grade base-portrait recipes plus the full script→TTS→avatar→B-roll pipeline. Use when the user wants a digital human to sell a product on TikTok Shop, | |
| 电商爆款主图 Ecommerce Main Image 电商爆款主图生成与编辑。上传已批准的商品图，生成搜索列表里能被认出、商品事实准确、卖点有画面证据、符合渠道规则的主图候选，并按单变量原则产出 A/B 测试组。覆盖白底基线图、可视差异图、内容电商场景图、配色 SKU 组、单变量背景测试五种场景，适用于淘宝天猫京东拼多多抖音小红书 Amazon TikTok Shop Shopify 的主图与首图。当用户需要「做主图」「爆款主图」「白底图」「主图测款」「测图」「A/B 测试主图」「一个 SKU 铺多个配色」时使用本技能。转化受价 | |
| 电商视频生成 Ecommerce Video ecommerce video, product video, shopping ad video, ecommerce video generator — turn a product (photos or link) into a conversion-focused ecommerce ad video. Use when the user has a product and wants a store, TikTok, or ad video. | |
| 多人对话配音 ElevenLabs Dialogue ElevenLabs eleven_v3 multi-voice dialogue: assign a different voice per line (up to 10) and render the whole conversation in one shot. Supports audio tags like [giggling], [whispers] — great for character dialogue, podcasts, and short skits | |
| 音乐生成 ElevenLabs Music ElevenLabs music_v1 model — generates 10–300s original music from a natural-language prompt. Good for BGM, ads, and short-video soundtracks. ElevenLabs music_v1 音乐生成模型，根据自然语言提示生成 10–300 秒原创音乐。适合 BGM、广告配乐与短视频音轨。 | |
| 音色库搜索 ElevenLabs Voice Search Search the ElevenLabs voice library by keyword, source, and category. Returns a playable preview for each matched voice so you can pick the right one before running TTS. 搜索 ElevenLabs 人声库：按关键词、来源、分类筛选可用音色，返回每个音色的试听样本，便于挑选后用于 TTS 配音。 | |
| 音效生成 ElevenLabs SFX ElevenLabs text-to-sound model — generates 1–22s short sound effects from a description. Suitable for foley, ambience, alerts, and game SFX. ElevenLabs 文本生音效模型，根据描述生成 1–22 秒短音效。适合拟音、环境声、提示音与游戏音效。 | |
| 语音转文字 ElevenLabs STT ElevenLabs scribe_v1 speech-to-text with auto language detection and optional speaker diarization. Suitable for subtitles, transcription, and meeting notes. ElevenLabs scribe_v1 语音转文字，支持自动语种识别与说话人分离，适合字幕、转录与会议记录。 | |
| 语音合成 ElevenLabs TTS ElevenLabs eleven_v3 text-to-speech with 12 curated multilingual voices and stability/similarity/style controls. Great for dubbing, audiobooks, and character dialog. Before picking a voice, you can search for the right one via elevenlabs-se | |
| 声音克隆 ElevenLabs Voice Clone ElevenLabs Instant Voice Cloning (IVC). Upload a clean voice sample to clone a custom voice usable with ElevenLabs TTS. ElevenLabs 即时音色克隆（IVC），上传一段干净人声样本即可复刻自定义音色，可用于 ElevenLabs TTS 配音。 | |
| 解说视频生成 Explainer Video explainer video, explainer video generator, animated explainer, training video — turn a document, topic, or brief into a narrated explainer video: outline, storyboard, voiceover, build, validate. Use when the user wants an explainer, course | |
| 文件转视频 File to Video ppt to video, word to video, excel to video, pdf to video, document to video — parse, outline, storyboard, voiceover, build, validate. Use when the user gives a document and wants an explainer, report broadcast, courseware, or training vide | |
| 录音转写 Fun ASR Alibaba Bailian Fun-ASR recording transcription. Supports Chinese, English and other languages, with auto language detection and speaker diarization. Suitable for subtitles, transcription, and meeting notes. 阿里云百炼 Fun-ASR 录音文件识别，支持中英文及多语种，自 | |
| 语音合成 Gemini 2.5 TTS Generate multilingual, highly natural audio using Gemini 2.5 text-to-speech. 使用 Gemini 2.5 强大的文本转语音能力，生成多语言、高自然度的音频。 | |
| 多模态大模型 Gemini 3.1 Google's multimodal model with strong long-context and vision understanding. Suitable for document parsing, image/video understanding, and structured output.... | |
| 全能生成 Generate A comprehensive generation skill. Can generate images, videos, and audio by automatically selecting the appropriate dlazy CLI model. 综合生成技能。能够根据用户意图自动选择合适的 dlazy CLI 模型来生成图片、视频或音频。 | |
| 通用大模型 GPT 5.5 OpenAI's general-purpose model balancing reasoning quality with response speed. Suitable for chat, writing, planning, and multimodal analysis. OpenAI 通用大模型，平... | |
| 图像生成 GPT Image 2 GPT Image 2 model for text-to-image and image editing. Supports generating images from text as well as editing and synthesizing images with reference inputs. GPT Image 2 图像生成与编辑模型。支持文生图，以及通过提供参考图进行图像编辑和合成。 | |
| 图像生成 GPT Image 2.5 Flare GPT Image 2.5 Flare, the speed tier of OpenAI's GPT Image 2.5 series, for text-to-image and image editing. Fast and cost-effective for high-volume everyday generation; supports reference images, aspect ratios up to 21:9 and quality up to ma | |
| 图像生成 GPT Image 2.5 Sunburst GPT Image 2.5 Sunburst, the precision tier of OpenAI's GPT Image 2.5 series, for text-to-image and image editing. Built for detailed creative work where fine detail, in-image text and composition matter; supports reference images, aspect ra | |
| 文本生成 Grok 4.2 Efficient text generation, dialogue QA, and logical reasoning using Grok 4.2 text model. 使用 Grok 4.2 文本大模型，进行高效的文本生成、对话问答与逻辑推理。 | |
| 视频生成 快马 Happyhorse 1.0 Happy Horse 1.0 video model — one model covers text-to-video (t2v), first-frame-to-video (i2v), reference-to-video (r2v), and video editing (edit). The selected mode is automatically routed to the matching sub-model. Happy Horse 1.0 视频模型，一站 | |
| 对口型 HeyGen Lipsync Speed HeyGen Lipsync Speed: Fast lip-sync model, ideal for scenarios requiring rapid generation. HeyGen Lipsync Speed：快速唇形同步模型，适合对生成速度要求较高的场景 | |
| 创意转视频 Idea to Video Turn a user's idea into the full pipeline: **story → characters → 3-view portraits → scenes → shots → keyframes → shot videos → concat**. First emit a **plan... | |
| 亚马逊商品图套件 Amazon Product Image Suite A professional product image generation skill purpose-built for the Amazon e-commerce platform. Outputs comply with Amazon's image guidelines while optimizing for click-through and conversio | |
| 图片生成 Image Generate Image generation skill. Automatically selects the best dlazy CLI image model based on the prompt. 图片生成技能。根据提示词自动选择最佳的 dlazy CLI 图片生成模型。 | |
| AI 生图选型指南 dLazy Image Guide Pick the right dLazy image model and get it right on the first call. Covers all 22 image tools with their prompt caps, size formats, reference-image support, and credit costs, plus editing and post-processing chains. | |
| 营销宣传册设计 Marketing Brochure A complete workflow skill for marketing brochure design, covering everything from requirements gathering, layout design, to mock-up delivery. It uses a 'layout-first + mandatory confirmation | |
| 图片仿制 Image Replicate Image replicate tool: analyzes the visuals, composition, colors, lighting, and style of the source image, builds a replicate prompt, and hands it off to Seedream 4.5 to generate a new image in the same style. 图像仿写工具：分析原图的视觉效果、构图、色彩、光影和风格，生成 | |
| 社交轮播图设计 Social Carousel A structured workflow skill dedicated to social-media carousel design. The core method is 'decide intent first, then execute,' using a 'single-confirmation + cover-first' two-phase flow. | |
| 社交媒体配图设计 Social Media Design A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs... | |
| 分镜图工作流 Image Storyboard A professional storyboard skill for film, advertising, short video, and educational narrative scenarios, built around a strict 'plan first, render later' flow. | |
| 图片抠图 Image Segmentation Image matting tool: separates foreground from background and returns transparent background URL, suitable for product image processing, character cutout, and composition. 图像抠图工具：将前景与背景分离并返回透明背景的 URL，适用于产品图片处理、人物抠图和合成。 | |
| 数字人动作视频 即梦 Dream Actor Convert static character images into vivid action videos with Jimeng Dream Actor. 使用即梦 (Jimeng) Dream Actor，将静态人物图片转化为生动的动作视频。 | |
| 图生视频 即梦首帧 Jimeng I2V Generate dynamic videos based on a single first frame image and prompts using Jimeng. 使用即梦 (Jimeng) 首帧生视频模型，基于单张首帧图片和提示词生成动态视频。 | |
| 首尾帧视频 即梦 Jimeng First-Tail Generate coherent transition videos using Jimeng's first and tail frame models. 使用即梦 (Jimeng) 首尾帧生视频模型，通过提供的第一帧和最后一帧图片生成连贯的过渡视频。 | |
| 数字人视频 即梦 OmniHuman 1.5 Generate realistic digital human broadcast videos from portrait images and audio/text using Jimeng OmniHuman 1.5. 使用即梦 (Jimeng) OmniHuman 1.5 模型，通过人像图片和音频/文本生成逼真的数字人播报视频。 | |
| 文生图 即梦 Jimeng T2I Text-to-image generation with Jimeng, quickly converting text to high-quality images. 使用即梦 (Jimeng) 进行文生图创作，快速将文字转化为高质量图像。 | |
| 音效生成 可灵 Kling SFX Generate matching scene sound effects based on text descriptions or video frames using Kling SFX. 使用可灵 (Kling) 音效模型，根据文字描述或视频画面智能生成匹配的场景音效。 | |
| 语音合成 可灵 Kling TTS Convert text into high-quality, emotional speech reading using Kling TTS. 使用可灵 (Kling) TTS 模型，将文本转化为高质量、情感丰富的语音朗读。 | |
| 文本生成 Kimi K3 Moonshot AI thinking model with text, image, and video understanding. Suited to complex analysis, coding, and writing that needs long reasoning chains. | |
| 声音克隆 可灵 Kling Audio Clone Generate customized speech that highly restores the timbre by uploading reference audio using Kling Audio Clone. 使用可灵 (Kling) 声音克隆模型，通过上传参考音频，生成高度还原该音色的定制语音。 | |
| 图像生成 可灵 Kling Image O1 Generate exquisite images with Kling o1 model. Supports text-to-image and image-to-image. 使用可灵 (Kling) o1 模型生成精美图像，支持文生图与参考图生图。 |
26,351–26,400 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.