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 |
|---|---|
| remote-skill-mcp congenital/remote_skill_mcp Dynamically exposes skill files from remote HTTP services as MCP tools, resources, and prompts, enabling clients to lazily load and read remote skill documentation via stdio. | |
| sdnext_mcp congenital/sdnext_mcp Enables LLM hosts to control SD.Next (Stable Diffusion Next) WebUI via 125 MCP tools for text-to-image, image-to-image, upscaling, control, tagging, model management, and more. | |
| spss-studio-mcp congjyu/spss-studio-mcp Enables AI agents to run IBM SPSS Statistics on macOS for statistical analyses, generate paper-ready charts, extract structured results, and execute mediation/moderation analyses with safety controls. | |
| Pavo 官网 MCP congliu1851-ui/pavo-website-mcp Enables local management of Pavo website accounts through a stdio MCP service, covering canvases, reference media, generation tasks, result downloads, and website assets. It also supports AI script episode planning plus character, scene, an | |
| Shopee MCP congminh1254/shopee-mcp Enables MCP clients to connect marketplace seller accounts, explore and call shop-level APIs, and manage orders, products, returns, inventory, pricing, and tracking with scoped read-only or read-write access. | |
| Knowledge MCP Server congtuuit/knowledge-mcp Enables AI agents to search, read, create, and maintain a local-first knowledge base using hybrid retrieval and Markdown note management through the Model Context Protocol. | |
| Mobile MCP Server congwa/mobile-agent A mobile automation MCP server providing 39 tools for Android and iOS device control, enabling AI agents to perform UI interactions, screenshot analysis, and coordinate-based operations. | |
| serverless-wiki-on-aws MCP Server coni524/serverless-wiki-on-aws MCP server that provides secure access to a serverless Markdown wiki on AWS, enabling AI assistants to read, search, and edit pages with space-level permission checks and Bedrock-powered semantic search. | |
| DevVault MCP conjureganja/devvault-mcp A local-first secrets vault for MCP clients that lets AI assistants use secrets by injecting them into child processes, without the plaintext ever entering the model's context. | |
| Desktop Bridge connbot/desktop-bridge A self-hosted MCP gateway that gives an AI client control of a real headed Chromium desktop and persistent workspace, with file, shell, process, and code-editing tools routed through a single control lease. Includes server-side read-only ob | |
| MCP Orders Server connect-abdullah/simple_orders_mcp A practice project that enables users to create and list orders within an in-memory system using MCP tools. It is designed to demonstrate how to integrate the Model Context Protocol with clients like Cursor using the FastMCP framework. | |
| PitchLense MCP connectaman/pitchlense-mcp Enables comprehensive AI-powered startup investment risk analysis across 9 categories including market, product, team, financial, customer, operational, competitive, legal, and exit risks. Provides structured risk assessments, peer benchmar | |
| GPTImage connected-mate/gptimage Image generation for Claude Code via ChatGPT subscription token. No API key needed. | |
| GPTVoice connected-mate/gptvoice A local MCP server that turns any text into narration, voice-overs, dialogues, audiobook chapters and per-shot video clips as MP3/WAV files — with emotions, acting modes, inline cues, presets and subtitles — driven by a ChatGPT sign-in rath | |
| iMessage Connector connected-mate/imessage-connector An MCP server that connects coding agents to iMessage on macOS, enabling reading, sending, searching, and reacting to messages under a configurable policy system. | |
| Bi-Temporal Knowledge Graph MCP Server connectimtiazh/graphiti-knowledge-mcp-server-starter Gives AI agents persistent memory with bi-temporal tracking, automatically extracting entities from natural language and enabling time-travel queries to understand facts as they existed at any point in history. | |
| Bi-Temporal Knowledge Graph MCP Server connectimtiazh/temporal-knowledge-graph-mcp-server Combines a bi-temporal knowledge graph with dynamic automation tool generation for AI agents, enabling persistent memory with temporal tracking and custom tool creation. | |
| mcp-server-entra-example connextai/mcp-server-entra-example A minimal, well-commented MCP server that authenticates its callers with Microsoft Entra ID (Azure AD). | |
| mcp-server-example connextai/mcp-server-example Minimal template MCP server with OAuth login and example tools for building custom servers on the Connext platform. | |
| mcp-server-product-studio connextai/mcp-server-product-studio A full-feature MCP server that serves as a product-development copilot, providing RAG search, SQL queries, and inline chart visualizations from seeded product data, with per-user identity and its own OAuth login. | |
| Connextify Official MCP Server connextify-com/connextify-mcp-server Enables AI assistants and agents to create and manage branded short links, dynamic QR codes, Link-in-Bio profiles, responsive landing pages, domains, campaigns, A/B tests, and marketing analytics through natural language. | |
| MCP+ connor9220/freecad-mcp-plus Lets AI assistants work inside a live FreeCAD session, building and inspecting models, running Python, driving the GUI the way a user would, and reviewing CAM jobs such as path statistics, posting, and G-code diffs. | |
| Unrestricted Development MCP Server connorboetig-dev/mcp2 Provides unrestricted access to your development environment with filesystem operations and shell command execution capabilities, including sudo support for local development machines. | |
| MSSQL MCP Reader connorbritain/mssql-mcp-reader Read-only Model Context Protocol server for Microsoft SQL Server, enabling safe schema discovery, profiling, and querying with zero risk of data modification. | |
| mssql-mcp-server connorbritain/mssql-mcp-server Enterprise-grade MCP server for Microsoft SQL Server, enabling semantic schema discovery, table profiling, safe data operations with preview/confirm, and multi-environment support. | |
| Kith connorhriley/kith Enables cooperating coding agents to coordinate locally by claiming task scopes, working in isolated Git worktrees, reserving shared resources like ports and serialized commands, and verifying changes before integration. Coordination state | |
| Antigravity MCP Bridge connormxy/antigravity-mcp-bridge Enables AI clients to run Google Antigravity coding sub-agents as MCP tools, with synchronous and asynchronous execution, task polling, and sandboxed Docker workspace isolation. | |
| pai connortessaro/pai Enables AI agents to manage API keys and child keys with budgets and spending caps, set plans and model routing, handle wallet payments and top-ups, and verify signed receipts through MCP tools. | |
| Connxio MCP Server connxio/cli Provides tools for managing Connxio integrations, code components, environment variables, and security configs through the Connxio v2 management API. | |
| avoid-ai-writing-mcp conorbronsdon/avoid-ai-writing-mcp Provides local, deterministic scoring and detailed auditing of text for AI-writing patterns via two read-only tools, with no network calls or language models. | |
| yt-analytics-mcp conorbronsdon/yt-analytics-mcp Owner-side YouTube analytics for AI agents: watch time, traffic sources, audience retention, playlist and podcast-series metrics, and episode-over-episode comparison. Read-only, all of it. | |
| langroid-captcha-solver-mcp-tool conorirish/langroid-captcha-solver-mcp-tool Enables Langroid agents to solve CAPTCHAs through CapSolver in authorized workflows, with structured validation, single-attempt limits, and human review for unsafe or incomplete requests. | |
| obscura-capsolver-tool conorirish/obscura-capsolver-tool Enables authorized, bounded CAPTCHA solving alongside Obscura browser sessions, with session binding, attempt limits, and human review fallback. | |
| Context Pods conorluddy/contextpods Context-Pods is a comprehensive development framework for creating, testing, and managing Model Context Protocol (MCP) servers. It provides a Meta-MCP Server that can generate other MCP servers through natural language descriptions or by wr | |
| XC-MCP: XCode CLI wrapper conorluddy/xc-mcp Problem: MCP clients can't efficiently use Xcode CLI tools because the build and simulator listing commands return more than 50k tokens, exceeding MCP limits. Solution: MCP tooling using progressive disclosure with intelligent caching. | |
| Beatport MCP API conorwd/beatport-mcp Enables searching for Beatport artists by name, returning structured data like names, IDs, and URLs. | |
| neuron-inspector conquext/neuron-inspector Turns Chrome into a local MCP server, giving AI agents 51 browser tools to inspect DOM, automate clicks, run JavaScript, audit security and accessibility, mock APIs, extract data, record demos, and more from any MCP client. | |
| Neuron MCP Server conquext/neuron-mcp-server Enables AI assistants to manage WhatsApp business operations including chatbots, broadcasts, campaigns, and contacts through 120+ MCP tools. | |
| grok-bridge conrad85/grok-bridge Enables ChatGPT Desktop and Codex to delegate substantial work to a locally installed Grok Build agent. Supports consultations, background builder/tester jobs, cancellation, session discovery, and transcript export. | |
| Remote PostgreSQL Database MCP Server conraddiao/remote-mcp-server-with-auth Enables you to chat with your PostgreSQL database via natural language, with GitHub OAuth authentication and role-based access control for secure read/write operations. | |
| claude-peers conradium-gif/claude-peers-v2 Lets Claude Code instances discover and message each other across sessions, with reliable delivery via hooks instead of experimental channels. | |
| browsermcp conradkoh/browsermcp Fork of Browser MCP with improved reliability, logging, and exit code handling; enables browser automation through MCP protocol. | |
| ProactiveAgent conradlu2740/proactiveagent Provides shared long-term memory and proactive suggestions across AI coding agents, enabling persistent user preferences and behavior rules that are remembered and applied across different tools. Includes memory capture, recall, extraction, | |
| jiandaoyun-mcp conred-lin/jiandaoyun-mcp MCP server that wraps Jiandaoyun (简道云) form operations, providing 9 tools for app discovery, form exploration, and data CRUD via SSE transport. | |
| Consail MCP consail-labs/consail-mcp Connects agents to Consail over local stdio, providing read-only, allowlisted tools to inspect environments, catalogs, datasets, actions, and validate pipelines without exposing database credentials or secret APIs. | |
| Posthouse conscienceli/posthouse Provides a shared, versioned workspace of plain files that Claude, ChatGPT, Gemini and CLI agents connect to over MCP with OAuth, so any agent can read and write the same project files, browse and restore git-backed history, and log dated h | |
| wow-site-mcp consecrated-hammer/wow-site Enables querying World of Warcraft character equipment, talents, profile, achievements, and realm lists via Blizzard's API, backed by shared caches and audit logging. | |
| hivemind-mcp consecrating/sanctify-hivemind Enables AI assistants to run WordPress site reconnaissance, malware/CVE security scans, SEO-spam/cloaking detection, and end-to-end natural-language goals through a stdio MCP interface, with the target site configured via environment variab | |
| consensa-mcp consensa-algo/consensa-mcp An MCP server that records consented, attributed, paid-for receipts for open-source dependencies, settling per-call USDC payments on Algorand with on-chain attribution. | |
| BuyICT MCP Server consentirdev/buyict.mcp Provides access to Australian Government ICT procurement opportunities from BuyICT, enabling users to search contracts across multiple marketplaces, view detailed opportunity information, and explore available procurement channels. |
40,201–40,250 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.