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 |
|---|---|
| svc-mcp-server cu0001/svc-mcp-server Enables AI assistants to interact with IBM SVC storage systems, checking system status and executing arbitrary SVC CLI commands via SSH. | |
| Cube MCP Server cubedevinc/cube-mcp-server Enables users to interact with Cube's AI agent for real-time analytics and data exploration through a chat interface. It supports generating data visualizations, performing SQL queries, and analyzing trends using natural language. | |
| mcp-image-tools cuber-it/mcp-image-tools Enables AI assistants to read, describe, resize, convert, and write images through the Model Context Protocol. | |
| mcp-playwright-tools cuber-it/mcp-playwright-tools Playwright browser automation tools for MCP enabling navigation, interaction, content extraction, and semantic locators. | |
| mcp_shell_tools cuber-it/mcp-shell-tools A full-featured MCP server for local development with filesystem, shell, editor, session persistence, and security features. | |
| Mailbunker_MCP cubetribe/mailbunker_mcp Zero-trust encrypted email archiving and search MCP server with real-time IMAP IDLE ingestion, AES-256-GCM encryption, SQLite FTS5 full-text search, and Obsidian vault export, exposing tools for AI assistants to search, retrieve, and manage | |
| YouTube MCP Extended cubetribe/youtube_mcp_over_api Manages YouTube channels via Claude Desktop, enabling AI-powered metadata optimization, video scheduling, playlist management, and backups with safety guardrails. | |
| CorelDRAW Live MCP cubicdev1/corel-mcp Enables control of an already open CorelDRAW document via MCP, with live editing, atomic undo, and safe coordinate-based operations. Supports Windows-only COM automation and integrates with desktop AI clients like Claude Desktop. | |
| google-mcp-suite-network cubicecho/google-mcp-suite-docker A tiny TypeScript wrapper that exposes five stdio MCP servers (Gmail, Calendar, Sheets, Docs, Drive) as Streamable HTTP endpoints, enabling a single Docker host to serve all Google services to any MCP client. | |
| @cubicecho/graphql-mcp cubicecho/graphql-mcp Converts a GraphQL schema into MCP tools, allowing AI to discover and call GraphQL queries and mutations via natural language. | |
| MCP Actual cubicecho/mcp-actual An MCP server for Actual Budget that connects to your Actual sync server, opens one budget file, and exposes it to MCP clients over streamable HTTP or stdio. | |
| RAGDown cubicecho/mcp-ragdown Enables agents to run hybrid dense and BM25 search over a local folder of Markdown files, read and write notes, and trigger reindexing as the folder changes. It also injects the most relevant sections into each prompt automatically and runs | |
| MCP Skills Manager cubicecho/mcp-skills-manager A server that serves reusable markdown skills (instructions, playbooks, references) to agents, with a web UI for editing and organizing skills, profiles, and both HTTP and stdio transports. | |
| mcp-zeromem cubicecho/mcp-zeromem Enables agents to store conversational turns in a single SQLite store and recall relevant context through lexical, entity, dense, and recent views without an LLM in the recall loop. It exposes MCP tools for remembering, bulk ingest, recall, | |
| CubicJ MCP Search cubicj/cubicj-mcp-search Enables LLM clients to run Brave LLM Context web searches and fetch readable markdown extracts from URLs, including specialized site extraction, PDF text, raw mode, and pagination. | |
| @cubiczan/chp-mcp cubiczan/cubiczan-chp-mcp Enables MCP clients to enforce CHP Profile B spend gates, issue and verify signed tool-approval receipts, and audit denied or authorized finance actions. | |
| Cubiczan MCP Server cubiczan/cubiczan-mcp-server Connects Off Grid mobile app to Cubiczan's multi-agent intelligence, exposing agent orchestration as an OpenAI-compatible API endpoint. | |
| finance-engines-mcp cubiczan/finance-engines Provides AI agents with deterministic, offline finance tools for commodity margin analysis, loan covenant compliance, invoice auditing, AP exception classification, and five-day close readiness. | |
| Continuity cubiix3/continuity Enables agents to maintain persistent project context locally, with source-backed memory and structured handoffs for seamless handover between interchangeable agents. | |
| MBTA MCP Server cubismod/mbta-mcp Enables AI clients to access Boston's MBTA public transit data, including real-time predictions, schedules, route planning, and service alerts. | |
| shopify-multistore-mcp cubitrek/shopify-multistore-mcp Enables conversational Admin API read/write across multiple Shopify stores with a production write-gate for safety. | |
| freedcamp cubo3/c3-claudemcp-freedcamp Connects Claude to a project-management account so it can list and manage projects, tasks, comments, and files through eight REST API tools, including creating and updating tasks and adding comments. | |
| bvbrc-mcp-server cucinellclark/bvbrc-mcp MCP server for querying BV-BRC biological data, submitting analysis jobs (assembly, annotation, BLAST, etc.), and managing workspace files via natural language. | |
| TradingView MCP Bridge cucoco-bot/tradingview-mcp Connects AI assistants to your local TradingView Desktop app via Chrome DevTools Protocol for chart analysis, Pine Script development, and workflow automation while keeping all data local. | |
| nonkyc-mcp cuddy-five/nonkyc Enables MCP clients to retrieve NonKYC.io market data and apply technical-analysis algorithms, currently focusing on BTCB2. | |
| claude-computer-use-mcp cudgk/claude-computer-use-mcp Wraps OpenAI Codex's computer-use engine to provide MCP tools for Windows screen operations including screenshot, UI automation tree, and input simulation, enabling Claude Code to control the desktop. | |
| ffmpeg-mcp cudgk/ffmpeg-mcp Enables LLMs to safely execute ffmpeg and ffprobe operations by passing arguments as arrays, avoiding shell injection, with support for probe, convert, trim, concat, thumbnail, watermark, speed, loudnorm, and batch processing. | |
| http-mcp cudgk/http-mcp Enables LLMs to make HTTP requests with OAuth2, session cookies, retry logic, and cURL command generation, while providing security features like SSRF protection and TLS enforcement. | |
| Cuebook MCP Server cuebook-public/cuebook-cli Provides market intelligence including asset search, market state, cues, events, and paper trading via the Model Context Protocol. | |
| CueHQ MCP Server cuehqsocial/cuehq-mcp Enables agents to retrieve ranked, deduplicated social threads where people ask for products, get reply drafts in the user's voice, and manage cues by marking replies or skipping them without posting on the user's behalf. | |
| metabase-mcp-server cuenca-mx/metabase-mcp-server Enables interaction with Metabase through its API, allowing users to list cards, databases, collections, execute queries, and create cards or bookmarks. | |
| Brave Search MCP Server cuer-bot/bravesearchmcp Enables web search via Brave Search API with freshness filters, full page content fetching, and usage tracking. Designed for Claude integration. | |
| MCP PostgreSQL cuevacelis/mcp_postgres A read-only MCP server for PostgreSQL that enables safe database introspection and querying via natural language. | |
| WhatsApp MCP Platform cuevaio/normal Provides a Model Context Protocol server for WhatsApp, enabling session management, webhook normalization, and tenant-based message operations through a public API and private control plane. | |
| claude-text-lab-mcp cuibitlabs/claude-watermark-remover MCP server for Claude Text Lab, enabling fact-preserving regeneration of Claude-generated text through inspect, sanitize, audit, rewrite, and benchmark operations, with auditable local quality gates and an unverified provenance verdict. | |
| awesome-student-ai-skills cuic19053-hue/awesome-student-ai-skills AI agent workbench and MCP server featuring 35 specialized skills for university student competition proposals, research grants, and scholarship applications. | |
| FilePilot AI cuiheng511/filepilot-ai Local-first MCP server for safely searching, reading, summarizing, tagging, deduplicating, and organizing local files with scoped access, read-only defaults, and dry-run plans. | |
| ShortsMonkey MCP cuilinhao/youtube-mcp A read-only MCP server for discovering YouTube outliers, analyzing viral Shorts, and evaluating single-video performance using stored snapshots. | |
| Coding Tools MCP cuishihui66/coding-tools-mcp Gives any MCP-compatible AI chat or agent a safe, model-neutral coding runtime with file read/search, structured multi-file patches, command execution, interactive sessions, and git operations, all confined to a single workspace and gated b | |
| MailMesh MCP culpen90/mailmesh-mcp A private, single-user MCP server that unifies Gmail, Microsoft 365/Outlook, and IMAP mailboxes for LLMs to search and read emails live, without storing or caching mailbox contents. | |
| Unanet GovCon ERP MCP Server culstrup/unanet-mcp-server Enables natural language interaction with Unanet GovCon ERP for project management, time tracking, financial reporting, and contact management, turning complex navigation into simple questions. | |
| Adversarial REA cultistsid/adversarial-rea This MCP server lets an agent run local binary inspection on ELF, Mach-O, PE, APK, IPA, ASAR, WebAssembly, and firmware targets — pulling symbols, sections, classified strings, disassembly, and mitigation posture, and flagging dangerous imp | |
| adversarial-rea cultistsid/adversarial-rea-mcp Enables local, evidence-first triage of native binaries by extracting bounded symbols, strings, sections, call sites, gadget candidates, and mitigation signals as MCP tools for agent-guided manual review. It also provides metadata-only insp | |
| cowork-qa-mcp cultistsid/cowork-qa-mcp An MCP server that provides an LLM with a real Chromium browser via Playwright, recording every action taken toward a stated goal and producing a structured trace so the run can be audited for goal achievement. | |
| Mobile Xray MCP cultivx/mobile-xray-mcp Enables AI-powered screenshot capture and UX/UI analysis of iOS simulators and Android emulators directly from your IDE. | |
| Remote MCP Server cultrix-github/mcp-timeserver A Cloudflare Workers-based MCP server that enables tool integration with Claude AI through OAuth login, allowing users to extend Claude's capabilities with custom tools like mathematical operations. | |
| sam-memory cumabozkurt/super-agent-memory Enables any MCP-capable coding agent to share one persistent, token-frugal local SQLite memory across tools, so decisions and project facts saved in one agent are known to the others. It exposes a tiny four-tool surface for searching and re | |
| Agent Deck MCP Server cumulativewebinc/agent-deck-mcp Enables MCP-capable agents to query a read-only 18-SKU product registry and verified 24-track catalog, with evidence-tied lookups, momentum scoring, skin configurations, and a hash-chain-verified ledger over stdio. | |
| cwi-mcp-public cumulativewebinc/cwi-mcp-public Provides read-only tools for searching CWI's catalog by title, artist, or ISRC, retrieving game metadata and play links, and querying ledger state, trust verdicts, and needle-drop verification via a streamable HTTP MCP server with Bearer to | |
| Cumulocity MCP Server cumulocity-iot/cumulocity-mcp-server Enables interaction with Cumulocity IoT platform for device management, measurements, and alarm monitoring. Supports filtering devices, retrieving measurements with time-based queries, and accessing active alarms through natural language. |
41,101–41,150 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.