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 |
|---|---|
| mcp-resource-server curtismu7/mcp-resource-server Serves read-mostly MCP tools across ten mock verticals and investment use cases, each backed by bundled SQLite databases, with optional proxying of investment tools to a user-supplied banking API. It enables clients to query realistic verti | |
| MultiBaas MCP Server curvegrid/multibaas-mcp-poc Enables AI assistants to interact with blockchain data and smart contracts through Curvegrid's MultiBaas platform, providing seamless access to blockchain operations and contract functionality. | |
| LINE Local MCP curzer1995-777/line-local-mcp A read-only MCP server that lets AI search and read both sides of a user's own LINE Desktop chat history on macOS using local database snapshots, without needing a LINE Official Account or altering unread state. | |
| LINE Local MCP curzer1995-777/line-local-mcp-user-key A read-only MCP server that lets AI search a user's own LINE Desktop chat history on macOS, providing tools to list chats, retrieve messages, and search conversations directly from the local encrypted database. | |
| smart-connections-mcp cusster/smart-connections-mcp Gives Claude Code read-only semantic search over an Obsidian vault by reading the embedding files the Smart Connections plugin already produces, so no API key or network is needed at query time and it works across the WSL/Windows boundary. | |
| helpdesk-workflow cust0diodev/helpdesk-mcp Enables natural-language queries to Tactical RMM, GLPI/TurboDesk, Zabbix, and Grafana for IT alerts, agents, tickets, problems, and dashboard metadata through a local MCP stdio server. | |
| YouTube Content Extractor MCP custardcream98/youtube-scrap-mcp Extracts YouTube video metadata, titles, and descriptions along with transcripts generated from subtitles or OpenAI Whisper speech-to-text. This server enables users to retrieve and analyze detailed video content directly within MCP-compati | |
| Custify MCP Server custifyofficial/custify-mcp Connects AI assistants to Custify customer success data for querying account metrics, health scores, and usage trends. It also supports taking actions such as creating notes, managing tasks, and triggering playbooks. | |
| SAP Sandbox MCP Server custodioanthonya/sap-mcp-server Connects Claude to SAP's public S/4HANA Cloud sandbox via MCP, enabling natural-language queries against real SAP OData structures like sales and purchase orders. | |
| amazon-mcp customcoder245/amazon_mcp Enables MCP clients to access Amazon Selling Partner and Advertising APIs for orders, inventory, pricing, ads, reports, and related read paths. | |
| Yonote MCP Server cutalion/yonote-mcp Provides API tools to interact with Yonote documents and collections, serving as an alternative to Notion with capabilities to list documents/collections and retrieve detailed document information. | |
| DS Web Local cute-aaa/ds-web-local Pluggable local agent capability hub that gives web-based DeepSeek local Agent capabilities (MCP/file operations/skills/desktop automation) and lets local agents reuse the same tools. | |
| OmniParser MCP cute-aaa/omniparser-mcp MCP server that wraps Microsoft OmniParser to provide screen parsing capabilities, returning structured UI elements (text, icons, coordinates) from screenshots for GUI agents. | |
| GO2 Sport MCP Server cutecar384/go2_sport_mcp Enables controlling a Unitree GO2 robot's motion (forward, backward, left, right, stop, stand up, and move with custom velocities) via MCP tools, using ROS2 commands. | |
| TrendRadar MCP Server cutejhy-code/trendradar Enables quick deployment of a trending news assistant with multi-channel notification and AI analysis, helping users stay updated with relevant news. | |
| kenshape-mcp cutetechguy/kenshape-mcp Enables AI assistants to create low-poly 3D art for Kenney Shape by drawing on a tile canvas, previewing extruded models, and saving projects or exporting GLB, OBJ, SVG, and PNG files. | |
| mcp-arr-lite cuti27/mcp-arr-lite MCP server for managing media downloads with Radarr and Sonarr, providing granular tools for searching, adding, and grabbing movies and TV shows through natural language. | |
| mcp-portainer-lite cuti27/mcp-portainer-lite Lightweight MCP server for Portainer CE, focused on Docker Swarm environments with read-only tools by default and optional mutations. | |
| hybrid-recall cutlerbenjamin1-cmd/hybrid-recall A local MCP server enabling hybrid search over documents, memory, and knowledge graphs for retrieval-augmented generation, with tools for SQLite, semantic memory, and entity-relationship queries. | |
| mcp-http-bridge cutlerbenjamin1-cmd/mcp-http-bridge This server bridges a stdio MCP server to HTTP, allowing MCP clients that communicate over HTTP to use the server's tools. It includes a per-tool allow/deny filter for security. | |
| procmem cutlerbenjamin1-cmd/procmem An MCP server for Windows process memory inspection/editing and crash dump analysis, enabling live memory scanning, patching, pointer resolution, disassembly, and offline dump examination via structured tool calls. | |
| shell-0 cutlerbenjamin1-cmd/shell-0 Provides direct, unsandboxed local machine access via filesystem, Python, Node.js, and shell commands for MCP agents. | |
| web-explorer cutlerbenjamin1-cmd/web-explorer MCP server providing free, IP-direct tools for web search, crawling, anti-detect browsing, and structured data from Wikipedia, Wikidata, GitHub, and YouTube. | |
| Airtable MCP cutolu2021/airtable_mcp Enables comprehensive interaction with the Airtable Web API to manage bases, tables, and records. It supports advanced features like batch record operations, CSV import/export, attachment uploads, and webhook management. | |
| CAPTCHA-MCP cuttlefish-1997/captcha-mcp Enables AI assistants to control a real Chrome browser with trusted mouse input to solve interactive CAPTCHAs like sliders, using a Chrome extension and OS-level mouse automation. | |
| game-art-mcp cuvara/game-art-mcp Provides AI agents with tools to manage and enforce consistent pixel-art style for 2D RPG games, including style definitions, asset registry, art memory, QA checks, provider execution, and versioning. | |
| comfy-mcp cuzelac/comfy-mcp MCP server that enables AI agents to control a local ComfyUI instance for image generation, allowing workflow understanding, parameter modification, execution, and model discovery. | |
| marketua cuzin85/marketua Search Ukrainian marketplaces (OLX, Hotline, Prom; Rozetka via Bright Data MCP) from a plain-language prompt and get structured offers with prices and links. | |
| Lark MCP cv-cat/larkagentx A Model Context Protocol server that enables AI models to perform function calls through Feishu/Lark messaging platform, using your personal account (no bot configuration needed) to create a full-featured AI assistant. | |
| hypercontext cv-superding/hypercontext-mode Enables AI coding agents to execute tools in a sandbox, retain searchable persistent memory, and reversibly compress long conversation histories through a unified MCP endpoint. | |
| mcp-browser cv3inx/browsermcp Self-hosted MCP server for browser automation with stealth patches, offering 50 tools for navigation, scraping, and session configuration. Runs via Docker and supports remote control from any MCP client. | |
| guerrillamail-mcp cv3inx/guerrillamail-mcp MCP server for the Guerrilla Mail temporary/disposable email API. Spin up throwaway inboxes, poll for mail, and read messages from any MCP client. | |
| dsh-web-mcp cv588888888888888888888ju/dsh-web-mcp An MCP bridge that exposes DeepSeek Harness (DSH) web UI's cordis RPC API as stdio tools, enabling any MCP client to manage workspaces, create/resume sessions, send messages, and fetch session stats. | |
| geocodage-mcp cvagner/geocodage-mcp MCP server for geocoding French addresses, POIs, and cadastral parcels using the Géoplateforme API. Supports direct geocoding, reverse geocoding, and capability queries. | |
| project-code-intelligence cvandesande/project-code-intelligence Hardware-accelerated codebase mapping that indexes Git repositories into Postgres/pgvector and serves code search, relationships, and static analysis results via a stdio MCP server. | |
| moa-mcp cvanes/mom-mcp A local stdio MCP server for running single-turn prompts across multiple model providers (OpenAI, Ollama, OpenRouter), allowing agents to fan out the same prompt to several models and combine the output. | |
| monzo-mcp cvanes/monzo-mcp Enables interaction with Monzo bank accounts for balance checking, transaction management, pot operations, and reconciliation through natural language. | |
| AWS AppRunner MCP Server cvanputt/mcp-example A boilerplate TypeScript MCP server with Express.js designed for AWS AppRunner deployment. Provides sample tools, resources, and prompts with Docker containerization and GitHub Actions CI/CD workflow. | |
| cveasy-mcp cveasy/cveasy-mcp Provides MCP clients like Claude with tools to query and manage security scans, CVEs, assets, findings, threat intel, and generate polished reports by acting as a lightweight adapter over the CVEasy backend API. | |
| cvefeed-mcp cvefeed/cvefeed-mcp MCP server providing vulnerability intelligence from CVEFeed.io, enabling CVE search, product subscriptions, and alert management via natural language. | |
| Polovnjak MCP cvele/polovnjak-mcp Enables agents to search and analyze used-vehicle listings across multiple Serbian marketplaces, including price statistics, history, and per-user saved-search watches. | |
| slimdown-mcp cventuresza-tech/slimdown Gives MCP clients such as Claude Code, Claude Desktop, Cursor or Codex a read_page tool that fetches a web page and returns only its main content as lean Markdown, stripped of navigation, ads and tracking. Accepts optional focus, links and | |
| EU Open Banking MCP cver-me/eu-open-banking-mcp Enables secure, read-only access to personal European bank accounts through Enable Banking and Cloudflare Workers, allowing MCP clients to list accounts, retrieve balances, search transactions, and summarize cash flow. | |
| MainWP MCP Server cvsloane/mainwp-mcp Enables AI assistants to manage WordPress sites via the MainWP Dashboard REST API, providing tools for site health monitoring, updates, plugin/theme management, and client/cost tracking. | |
| Proxell MCP Server cvsper/proxell-mcp Implements the Proxell Exchange Protocol (PXP) to enable AI agents to autonomously discover, price, negotiate, and trade digital assets like leads, datasets, and APIs. It provides a comprehensive suite of over 90 tools for managing the full | |
| YouGile MCP cvtmysxul/yougile-mcp An unofficial, security-focused MCP server for YouGile REST API, providing policy-controlled tools with read-only defaults and optional direct-write profiles. | |
| Packet-Tracer-MCP cvz-dev/packet-tracer-mcp Enables natural-language construction and configuration of Cisco Packet Tracer topologies, including devices, cabling, IOS commands, VLANs, routing protocols, and network reproduction from diagrams. | |
| Metabase MCP Server cw-codewalnut/metabase-mcp-server Connects Metabase business intelligence platform to AI assistants, enabling users to query data, create dashboards and charts, manage databases, and interact with BI assets using natural language through the Model Context Protocol. | |
| auto-chrome-mcp cw02326/auto-chrome-mcp An MCP server for controlling Google Chrome through native messaging, featuring forced reconnect, diagnostic tools, and automatic permission management. It enables AI assistants to automate browser tasks such as navigation, scraping, and in | |
| nudgeoss cw12574/nudgeoss MCP server that provides an ask_human tool so AI agents can request decisions from users via chat apps, with deadlines, defaults, and an append-only audit log. |
41,201–41,250 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.