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 |
|---|---|
| GCP BigQuery MCP Server erayguner/db-mcp Enterprise-grade MCP server for Google Cloud BigQuery with keyless Workload Identity Federation authentication, enabling secure SQL query execution, dataset management, and schema inspection with comprehensive audit logging and encryption. | |
| MCP Server ELK erbiantoo/mcpserver A production-oriented, read-only MCP server for secure ELK stack analysis, compatible with OpenClaw. | |
| KiCAD MCP Server ercbb/kicad-mcp-server Enables AI assistants like Claude to interact with KiCAD for PCB design automation, providing comprehensive tool schemas and real-time project state access. | |
| CustomFigmaMcp ercinakkaya/customfigmamcp Analyzes Figma design files through natural language queries, enabling component detection, color extraction, and consistency checks via AI assistants like Claude in Cursor. | |
| colab-mcp erdhirajpatel/colab-mcp An MCP server for bridging your local agent to a Colab session in the browser. | |
| Apple Dev MCP Server erdncyz/apple-dev-mcp-server Provides live Apple documentation fetching, Xcode build error analysis, and Swift Evolution proposal checking for Apple development assistance. | |
| claude-design-mcp erdnj/claude-design-mcp Unofficial stdio MCP proxy for Anthropic's Claude Design MCP. Enables using Claude Design with any MCP host like Cursor, Claude Code, or Codex. | |
| Vectorize MCP Server erdogan4423/vectorize-mcp-server Enables users to perform vector retrieval, extract and chunk documents into Markdown, and generate private deep research reports via configured pipelines. | |
| MS 365-21V MCP Server erdogan4885/ms-365-21v-mcp-server An MCP gateway that lets AI clients act as the signed-in user against 21Vianet-operated Microsoft 365 through Microsoft Graph China, enabling mail reading and sending, calendar scheduling and conflict checks, and OneDrive/SharePoint file se | |
| Meticulous Espresso MCP Server erdos2n/meticulous-mcp-server Connects your Meticulous espresso machine to an LLM, enabling recipe generation from natural language, shot analysis, grinder dial-in, and a persistent shot diary through Claude. | |
| Azure DevOps MCP Server for Cursor erealexandru/azure-devops-mcp-server-for-cursor Enables interaction with Azure DevOps through natural language in Cursor IDE. Supports work item management, pull requests, builds, releases, test management, and guided workflows for development teams, QA testers, and release management. | |
| imap-mcp erebetez/imap-mcp Provides email fetching and drafting capabilities via IMAP. | |
| rigol-mcp erebusnz/rigol-mcp An MCP server for controlling Rigol DS1000Z series oscilloscopes over LAN using natural language. It enables users to take measurements, configure instrument settings, and capture screenshots directly through an MCP client. | |
| Pixasso MCP Server erebuzzz/pixasso Orchestrates full-stack frontend engineering—from intent discovery and design genome generation to automated multi-viewport QA—through JSON-RPC tools for AI agents. | |
| Mandate Desk erekola/mandate-desk Enables owner-approved agent transfers: an owner sets limits, reviews requests, and approves exact contents while agents prepare and execute approved plans, including local simulation and unsigned Sepolia preview. | |
| turva-mcp erekola/turva-mcp Read-only MCP server that answers questions about turva.dev from its published data. Five tools return JSON: the service catalog with prices, contact and operator details, engagement principles and dated agent-readiness and security evidenc | |
| kendo-angular-unofficial-mcp eremesng/kendo-angular-unofficial-mcp Enables AI assistants to browse, search, and extract code examples from the public Kendo UI for Angular documentation, providing an accessible alternative to official premium MCPs. | |
| Thoth-Mem eremesng/thoth-mem Persistent memory for AI coding agents, storing learned architecture decisions, patterns, and bug fixes in a local SQLite database with full-text search, enabling agents to recall information across sessions. | |
| nano-pdf-mcp eren0315/nano-pdf-mcp Memory-efficient MCP server for reading and splitting large PDF files page by page using PyMuPDF. | |
| attocode-code-intel eren23/attocode Enables MCP clients to understand codebases through AST navigation, keyword and optional embedding search, symbol and reference analysis, impact assessment, test suggestions, change review, and shared knowledge across local working trees or | |
| large-repo-mcp-server erenari/large-repo-mcp-server A stdio-based MCP server for fast, bounded search and navigation of large code repositories using ripgrep, with tools for project-wide search, symbol lookup, file listing, and code reading. | |
| huawei-ont-mcp erenn0989/huawei-ont-mcp Reads LAN devices and WAN status from a Huawei GPON ONT fiber modem. | |
| PitchLink MCP erenozdemirrr/pitchlink-mcp An MCP server that reads startup pitch drafts from Notion to provide comprehensive investor-style analysis and scoring. It evaluates key areas like market opportunity and team strength, delivering feedback through a visual dashboard. | |
| linux-diag-mcp erenozgu/linux-diag-mcp Lets an AI assistant inspect a Linux machine through ten fixed read-only tools covering load, memory, disks, processes, ports, interfaces, systemd units, the journal, and allowlisted log files. All output is size-capped, control-character-s | |
| ccg-mcp erenshkigal111/ccg-mcp Enables Claude Code Desktop to call OpenAI Codex/GPT-5 and Gemini APIs via stdio MCP, without requiring local CLI tools. | |
| NotebookLM MCP Server erenu97/notebooklm-mcp MCP server for Google NotebookLM that drives a real Chrome via Patchright so an agent can chat against a notebook, ingest sources, generate audio overviews, and read DOM-level citations. | |
| myst-mcp eresearchqut/myst-mcp Enables searching any MyST Markdown or Jupyter Book documentation site that publishes a search index, without crawling or embeddings. | |
| vps-mcp ereskovsky/vps-mcp MCP server for managing VPS servers via SSH, enabling command execution, file transfer, Docker management, and server documentation from within Claude. | |
| ahttp-mcp eretly/ahttp Enables LLM agents to open web pages, retrieve compact DOM-derived action snapshots, and execute server-authorized interactions with access modes and confirmation guardrails. | |
| forge-mcp erewhon/forge Provides MCP tools for AI-powered code review, research, and book writing via a fleet of coding agents. | |
| MAX MCP erezcohen85/max_mcp Enables pulling credit card expenses from MAX Israel, categorizing them, and exposing them as MCP tools for analysis and reporting. | |
| asset-finder-mcp erfangit23/asset-finder-mcp All-in-one asset search MCP server for AI coding agents, enabling search for icons, logos, stock photos, vectors, and emoji from a single tool call. | |
| MT5 MCP Server erfangit23/mt5-mcp-server Enables AI assistants to interact with MetaTrader 5 for market data, indicators, order placement, position management, and risk-based lot sizing, with dry-run safety on by default. | |
| AutoCAD MCP server erfouni/autocad-ai Enables AI assistants to draw and manipulate 2D designs in a running AutoCAD instance through MCP tools, exposed via HTTP and a tunnel for cloud clients like ChatGPT and Manus. | |
| OpenRouter Model Router MCP Server erfouni/connect-gpt-to-all-models-with-openrouter Enables ChatGPT to delegate tasks to various AI models (e.g., Gemini, Claude, Qwen) via OpenRouter, acting as an orchestrator that returns results with confirmed model usage. | |
| Mizito AI Connector erfouni/mizito-ai-connector Exposes 105 MCP tools that let Claude and ChatGPT read and analyse a Mizito workspace's chats, projects, tasks, calendar, letters, minutes, surveys, notes and files (including PDF, Word, Excel and image content). Write, CRM and admin tools | |
| CloudNet MCP Server ergo042/cloudnet-mcp Provides an interface to the CloudNet Service REST API v3, enabling AI assistants to observe and manage CloudNet nodes, services, and players. | |
| olx-aluguel-mcp ergocodes/olx-aluguel-mcp Enables users to search OLX Brasil rental listings from Claude, extract detailed payment, contact, address, and advertiser data, compare listings, and check market prices. It also provides heuristic fraud-risk scoring, safety checklists, an | |
| workana-mcp ergocodes/workana-mcp Enables read-only searching and evaluating public freelance projects, including detailed views, opportunity scoring, and best-opportunity lists based on user skills. | |
| Two Truths and a Twist ergodiclabs/twotruthsandatwist An MCP-based trivia game server where AI creates rounds with two true statements and one false 'twist' about various topics, letting players guess which statement is false. | |
| gmail-sendas ergofobe/gmail-sendas-connector Enables sending email from verified Gmail sendAs aliases and downloading inbound attachment data as files, filling gaps in the stock Gmail MCP. | |
| imogen ergofobe/imogen-server Enables AI assistants to search a self-hosted photo and video library, view photos, and manage albums through MCP, with all actions scoped to the connected account and vault items hidden. | |
| vision-eye ergou-yu/vision-eye Enables text-only LLMs to perceive images entirely on-device, providing vision capabilities like image description, OCR, table extraction, UI analysis, and region focusing without any cloud APIs or API keys. | |
| MCP Server Bootcamp ergrato-dev/bc-mcp-server A comprehensive free bootcamp that teaches the creation of Model Context Protocol (MCP) servers, covering everything from basics to advanced enterprise architectures in 7 weeks. | |
| ask-user ergunsh/ask-user-mcp-app An MCP App that enables AI agents to ask users multiple questions with tab-based navigation, multiple-choice options, multi-select support, and custom text input, all rendered inline in the conversation. | |
| MCP server for LogSeq ergut/mcp-logseq Interacts with LogSeq via its API. | |
| youtube-transcript-mcp ergut/youtube-transcript-mcp The first remote MCP server that enables Claude AI to extract transcripts from YouTube videos with zero-setup access on any platform, including mobile. | |
| HANAPerformanceMCP erhankeseli/hanaperformancemcp An MCP server that gives AI assistants direct access to SAP HANA performance data — query plans, execution stats, memory usage. | |
| Unreal Engine MCP Server erhansiraci/ue-mcp Enables AI assistants to interact with Unreal Engine via Remote Control API for actor, asset, level, and editor operations. | |
| thru-mcp-server erhnysr/thru-mcp-server A read-only Model Context Protocol server for the Thru blockchain alphanet, enabling account queries, transaction status, program events, name service resolution, faucet status, and network health via the official @thru/sdk. |
48,351–48,400 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.