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 |
|---|---|
| Twilio Microvisor MCP Server ag2-mcp-servers/twilio---microvisor An MCP server that enables communication with Twilio's Microvisor API, allowing interaction with IoT devices managed by Twilio's Microvisor through natural language. | |
| Twilio Routes MCP Server ag2-mcp-servers/twilio---routes An MCP server that enables interaction with Twilio's Routes service through natural language, allowing users to manage SIP domains, trunks, and call routing configurations. | |
| Twilio Verify MCP Server ag2-mcp-servers/twilio---verify An MCP server that enables interaction with Twilio's Verify API, allowing users to manage phone verification services through natural language commands. | |
| UCO Bank MCP Server ag2-mcp-servers/uco-bank Enables interaction with UCO Bank APIs through the Model Context Protocol, supporting banking operations such as account lookup, transaction history, and more. | |
| Velo Payments API MCP Server ag2-mcp-servers/velo-payments-apis An MCP server that enables interaction with Velo Payments APIs for global payment operations, automatically generated using AG2's MCP builder from the Velo Payments OpenAPI specification. | |
| Vimeo MCP Server ag2-mcp-servers/vimeo Provides a Multi-Agent Conversation Protocol server for interfacing with the Vimeo API, allowing agents to interact with Vimeo's video platform services through natural language. | |
| VTEX Headless CMS MCP Server ag2-mcp-servers/vtex-headless-cms An MCP Server that enables interaction with VTEX's Headless Content Management System API, allowing users to manage content through natural language commands. | |
| WhatsApp Business API MCP Server ag2-mcp-servers/whatsapp-business-api An MCP (Multi-Agent Conversation Protocol) Server that enables interaction with the WhatsApp Business API, allowing agents to send messages, manage media, and perform other WhatsApp business operations through natural language. | |
| Google Workflows MCP Server ag2-mcp-servers/workflows-api An MCP (Multi-Agent Conversation Protocol) Server that enables interaction with Google Workflows API, allowing management of workflow executions and definitions through natural language commands. | |
| Workload Manager MCP Server ag2-mcp-servers/workload-manager-api An MCP server that enables interaction with Google Cloud's Workload Manager API, allowing users to manage workloads through natural language commands. | |
| Xero OAuth 2 Identity Service API Server ag2-mcp-servers/xero-oauth-2-identity-service-api An MCP server that enables interaction with Xero's Identity Service API, providing OAuth 2 authentication and identity management capabilities through natural language commands. | |
| Zoom API MCP Server ag2-mcp-servers/zoom-api Enables interaction with Zoom services through the Zoom API. Provides access to Zoom's functionality for meeting management, user operations, and platform features through natural language. | |
| Verantyx Vera MCP Server ag3497120/verantyx-vera-beta Provides LLM agents with hallucination-free external memory and knowledge tools, including ask, remember, forget, math, and code reasoning, with deterministic, traceable outputs. | |
| anysoar MCP server ag3ntsc4rn/anysoar Enables AI agents to operate any SOAR platform through a single tool interface, translating operations to vendor-specific drivers while enforcing read/write/destructive action guardrails, approvals, audit logging, and a kill switch. | |
| Memory Bear ag4852/memory-bear An MCP server that transforms markdown notes into a searchable knowledge base with semantic search, smart note creation, and flashcard management for AI assistants. | |
| cmdshellmcp ag88/cmdshellmcp This MCP server enables AI agents to run a restricted set of Unix commands, manage files within a sandboxed working directory, apply patches, and fetch web content, providing controlled local automation without full shell access. | |
| screen-mcp-server aga134/screen-mcp-server Gives AI coding assistants real-time screen capture capabilities with multi-monitor support, PDF text extraction, and answer logging. | |
| bookstore-mcp-server aga87/mcp-server-and-ai-agent-bookstore An MCP server that exposes tools for querying a bookstore inventory, allowing AI agents to search and retrieve book information via the Model Context Protocol. | |
| medimgq agadin/medical-image-question-creator Enables an AI to generate labeled CT images with answer keys for medical-school practice questions, including click-the-structure hotspot masks that align with the images. | |
| CadPlot MCP agadraphy/cadplot-mcp An MCP server for repeatable AutoCAD plotting workflows, enabling safe inspection of DWG files, page-setup verification, and consistent PDF publishing. | |
| yt-dlp agaldemas/yt-dlp-mcp Enables video metadata extraction and downloading from platforms like TikTok and YouTube via yt-dlp, including direct MP4 links and local downloads with MP3 conversion. | |
| Git Worktrees for ZCode agallardol/zcode-git-worktrees Manages isolated Git worktrees for each ZCode thread, enabling safe parallel development with automatic setup, snapshot-based removal, PR review, and background agent orchestration. | |
| mcp-server-bwt agalliani/mcp-server-bwt Enables AI assistants to manage Bing Webmaster Tools, including site verification, URL submissions, traffic analytics, crawl monitoring, and keyword analysis through natural language. | |
| Vulnary-MCP agam-s/vulnary-mcp MCP server for querying CVE and package vulnerability data from NVD and OSV.dev, allowing CVE lookups and dependency scanning. | |
| dumbdown agammann/dumbdown Enables assistants to call dumbdown_explain via stdio to convert code and technical documentation into plain-language explanations with analogies, walkthroughs, and jargon definitions. | |
| Frankly Hotdog agammann/not-hotdog Classifies a user-selected image and returns a verdict of HOTDOG, NOT HOTDOG, or UNCERTAIN, using structured output that surfaces uncertainty and never mislabels an error as NOT HOTDOG. It is an entertainment-only tool and does not persist | |
| MCP CTF Challenge Server aganita/mcp-ctf-challenge A vulnerable MCP server designed for educational CTF challenges. It demonstrates various MCP security vulnerabilities in a controlled environment. | |
| MCP Server with OpenAI-Compatible Model Support agaonker/mcp-apps Provides tools for file system operations and text generation using OpenAI-compatible models via the Model Context Protocol, supporting both stdio and HTTP transports. | |
| http-mcp-server agarcia-seidor/http-mcp-server Mock HTTP server implementing MCP (Model Context Protocol) with streamable HTTP transport, health endpoint, and example tools like echo and time. | |
| Teams MCP agarcia996/teams-mcp Enables AI assistants to interact with Microsoft Teams and Microsoft Graph, covering chat and channel messaging, message search, user and team directory lookups, and file upload or download via hosted content. Supports device-code OAuth aut | |
| FastMCP OpenAPI agarwal-saurabh/multiapimcpserver Dynamically generates MCP tools from OpenAPI specifications, enabling AI assistants to interact with any REST API through natural language. Supports multiple APIs with authentication, parameter validation, and integration with Claude Deskto | |
| OpenTelemetry MCP Server agarwalvivek29/opentelemetry-mcp Enables AI agents to query Prometheus metrics and Loki logs for intelligent alert investigation and troubleshooting. Provides service discovery, metric querying, log searching, and correlation tools to help identify root causes of issues. | |
| quickwit-mcp agarwalvivek29/quickwit-mcp A read-only MCP server that exposes Quickwit log search and aggregations to LLM clients, enabling natural language log investigation. | |
| dreamDEX Perps MCP agasavenue17/dreamdex-perps-mcp Enables AI agents to read dreamDEX perpetual markets, inspect portfolio state, claim testnet credit, and place safety-guarded on-chain orders on the Somnia Hideki testnet through MCP tool calls. | |
| interp-playground agaskell/interp-playground Enables AI agents and humans to inspect a language model's internals through activations, sparse-autoencoder features, steering, and ablation, currently targeting Gemma 2 2B with Gemma Scope on Apple Silicon. The current server exposes run_ | |
| TrinityCore MCP Server agatho/trinitycore-mcp Enterprise-grade MCP server with 107 tools for TrinityCore bot development, game data access, AI code review, performance profiling, and a full web UI. | |
| queregalarme agaver2/queregalarme Enables AI agents to browse public wishlists, check available gifts within budget, reserve items, and draft coordination messages without scraping or clicking through the UI. | |
| memopaper-mcp agcraft/memopaper-mcp MCP server for memopaper.dev that gives AI agents direct access to a memo buffer, enabling memo creation, retrieval, editing, and deletion via natural language. | |
| vapi-mcp-server ageborn-dev/vapi-mcp-server Enables AI agents to manage Vapi Voice AI platform resources such as assistants, calls, phone numbers, analytics, and workflows through natural language. | |
| Blender MCP ageless-h/blender-mcp MCP server for AI-assisted Blender automation, providing layered tools for perception, declarative and imperative writing, and fallback operations. | |
| gog-bridge agelyhq/mcp-gog-bridge Enables Claude Desktop and Cowork to execute the full gog CLI for Google Workspace through natural language, with a policy that locks the account and blocks administrative or identity-changing commands. | |
| TradingView MCP Jackson agencd/tradingview-mcp Enables automated trading analysis and chart management through the TradingView Desktop app. Provides tools for morning brief workflows, Pine Script development, chart navigation, and replay mode practice with local data processing. | |
| UX Laws MCP agencia-tecnologica-multiverse-limitada/ux-ui-mcp An MCP server for auditing and analyzing user interfaces across 20 platforms based on 30 fundamental UX laws. It provides automated code detection, platform-specific checklists, and comparative analysis to optimize design usability. | |
| UsageWall agenciabakucl-coder/usagewall Live LLM pricing as an MCP server. Ask Claude or any MCP client 'how much does this prompt cost?' and get real numbers from a hand-checked pricing table for every major LLM. | |
| mcp-runrunit agencian1/mcp-runrunit MCP server for interacting with Runrun.it API, enabling task and comment management for project management workflows. | |
| Agent Skills API agency-skills/agency-skills-b A multi-agent system for code rules management using MCP, supporting search, generation, and architectural validation through specialized agents. | |
| CC Fig MCP agenisea/cc-fig-mcp Enables bidirectional control between Claude Code and Figma with real-time sync, allowing creation and manipulation of design elements, component management, and document inspection through natural language commands. | |
| Agent Billy MCP Server agent-billy/mcp Provides secure, role-based access to Stripe billing operations for AI assistants like Claude and Cursor. It enables users to manage customers, subscriptions, invoices, and refunds through a local proxy with server-side permission enforceme | |
| agent-chaperone agent-chaperone/agent-chaperone Screens AI agent tool calls before execution and tool results before the agent reads them, acting as an MCP proxy and hooks adapter with shadow, enforce, and strict policy modes. | |
| @agent-dispatch/mcp-server agent-dispatch/mcp-server Enables AI assistants to spawn and manage long-running cloud agents via MCP, offloading tasks to AWS Bedrock AgentCore. |
24,351–24,400 of 115,409
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.