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 |
|---|---|
| kr-pc-deals-mcp edward-kim-dev/kr-pc-deals-mcp Enables AI assistants to search and compare Korean PC parts prices from Danawa and Compuzone, build assembly estimates with automatic compatibility checks, and track price history. | |
| noggin edward-zion-saji/noggin Local-first knowledge base that ingests activity from Slack, GitHub, agent sessions, and CLI, stores provenance in SQLite, and exposes the brain via MCP, CLI, Slack, and dashboard for recall and skill proposals. | |
| Recall edwardaktas75763/recall Enables Claude and other AI agents to retain and retrieve decisions, patterns, and knowledge across sessions, surviving context limits, restarts, and compaction. It supports self-hosted Redis/Valkey or managed SaaS, with workspace-scoped me | |
| engram-serve edwardalexander9252/engram Serves focused context to AI coding agents from a locally indexed structural code graph of entities, Git relationships and recorded project memory, so relevant code and past mistakes can be retrieved before reading whole files. Queries can | |
| apollo-io-mcp-server edwardchoh/apollo-io-mcp-server Exposes Apollo.io API functionalities as MCP tools for people and organization enrichment, search, and job postings. Enables natural language interaction with Apollo.io data. | |
| Flightradar24 MCP Server (Cloud Run) edwardchuang/fr24-mcp-cloudrun Wraps the official Flightradar24 MCP Server to expose it over HTTP/SSE for remote access, enabling MCP-compatible clients to query flight tracking data with client-provided API keys. | |
| ExpoSnap edwarddjss/exposnap Enables AI assistants to view and analyze screenshots from React Native/Expo applications for AI-powered mobile UI development. Integrates with Claude, Cursor, VS Code and other MCP-compatible editors. | |
| business-central-mcp edwardhayes91/business-central-mcp Gives AI assistants direct access to Microsoft Dynamics 365 Business Central via the native WebSocket protocol, replacing OData, APIs, and browser automation. Enables page navigation, data reading/writing, actions, searches, and report exec | |
| Microsoft Business Central MCP Server edwardhayes91/mcp-business-central MCP server for Microsoft Dynamics 365 Business Central, enabling AI assistants to perform CRUD operations, query data, and retrieve schemas via Business Central API v2.0. | |
| MCP X++ Server edwardhayes91/mcp-xpp An MCP server for Microsoft Dynamics 365 Finance & Operations development, enabling object creation, modification, deletion, and analysis through the MCP standard. | |
| msw-mcp edwardhuh60/msw-mcp Enables AI assistants to dynamically add, update, and remove Mock Service Worker request handlers at runtime through MCP tools, with setup scaffolding and a WebSocket browser bridge. | |
| MyDocsMCP edwardmaster7/mydocsmcp MCP server that enables semantic search over local PDF collections using local RAG, with automatic indexing of new documents. | |
| Houtini LM edwardnovikov861/houtini-lm Enables Claude Code and other MCP clients to delegate bounded tasks such as code review, boilerplate generation, commit messages, and format conversion to a local or cloud OpenAI-compatible LLM. It keeps the main model as architect while cu | |
| Oh My Cassette edwardosipov88/oh-my-cassette Enables users to turn raw clips into finished, beat-synced montage videos through natural-language chat, including shot selection, music matching, subtitles, transitions, and picture-in-picture. It lets agents analyze footage, plan and revi | |
| mcp-image-reader edwardpro621/mcp-image-reader Enables AI agents to actually see images from local file paths and remote URLs by returning them as vision input, with automatic JPEG re-encoding, referer handling, and optional directory allowlisting. | |
| MathJS Calculator MCP Server edwardpwtsoi/mathjs-mcp Provides mathematical calculation capabilities using MathJS, enabling expression evaluation and derivative calculation. | |
| companies-house-mcp edwardradford/companies-house-mcp Enables read-only access to the UK Companies House register, letting models answer questions about company status, ownership, filings, charges, and officers via seven purpose-built tools. | |
| mcp-image edwardrorato6434/mcp-image Enables AI assistants to generate and edit images from natural-language prompts, with automatic prompt enhancement and quality presets across Gemini, OpenAI, and BytePlus Seedream. | |
| agent-stealth-browser edwardsaga/agent-stealth-browser A stealth-enhanced browser automation MCP server for AI agents to interact with websites while bypassing anti-bot detection mechanisms like Cloudflare and reCAPTCHA. | |
| Deep Code Reasoning MCP Server edwardschaefer24/deep-code-reasoning-mcp Enables Claude Code to escalate complex code-analysis tasks to Google Gemini for large-context execution tracing, cross-system impact analysis, performance bottleneck detection, and hypothesis testing. | |
| slop-guard edwardschaefer24/slop-guard Enables agents to analyze inline text or files for formulaic AI writing patterns, returning scores, bands, violations, and concrete advice through MCP tools. Supports custom rule configurations without external API calls. | |
| Azure HPC/AI MCP Server edwardsp/mcpserver Enables users to query Azure HPC/AI Kubernetes clusters for GPU node information and InfiniBand topology details through kubectl commands. Provides tools to list GPU pool nodes with their status and retrieve network topology labels for high | |
| jev-compactor edwardyen724-g/jev-compactor Provides deterministic context compaction and safety gating for AI agents, preserving original messages verbatim while dropping irrelevant ones and flagging destructive commands. | |
| job-tracker-mcp edwardyun12/job-tracker-mcp Enables users to manage job applications (add, list, update status) through natural language conversations, using Notion as the backend. | |
| Weather MCP Server edwiin1688/claudelocalmcp.ts Enables Claude to query weather information for a specified area using the OpenWeatherMap API. | |
| Awareness Local edwin-hao-ai/awareness-local A local-first MCP memory server that gives AI coding agents persistent memory with hybrid semantic and keyword retrieval, working fully offline. | |
| Background Browser MCP edwin-hou/background-browser-mcp Gives Codex and, optionally, ChatGPT access to a separate signed-in browser on your Windows PC, so saved website logins can be reused and routine browsing runs headless. Sessions are signed in once per site through a local dashboard, with s | |
| dgcp-mcp-server edwin042331-hue/dgcp-mcp-server Connects AI assistants and agents with the Dominican Republic's public procurement transactional portal and open data API, enabling searches of tenders, contracts, bids, and suppliers, plus legal threshold calculations, proposal drafting, s | |
| MCP Data Analytics Server edwin1719/mcp-data-analytics-server A FastMCP-based server that provides data analysis tools for processing, analyzing, and visualizing data with an intuitive Streamlit web interface. | |
| weather edwin45168899/claudelocalmcp.ts Enables Claude to fetch real-time weather data for any city using the OpenWeatherMap API. | |
| Power BI China (21V) MCP Server edwindigital/powerbi-cn-skill Enables read-only analysis of Power BI China reports and semantic models through the China REST API, supporting report discovery, metadata retrieval, and DAX query execution. | |
| mcp-witness edwiniac/mcp-witness Cryptographic proof of every AI decision. An immutable, verifiable audit trail MCP server. | |
| mcp-rag-assistant edwinin01/mcp-rag-assistant Provides RAG-based knowledge retrieval and document management as MCP tools, supporting hybrid search, reranking, and retrieval process visualization. | |
| Berry Brain edwinmurari/berry-brain Enables AI clients to save checked outcomes and reuse tested lessons as private project memory across Codex, Claude Code, and other stdio MCP clients. | |
| jev-kit edwinzhu68-ops/jev-codex-kit Enables coding assistants to gather bounded source evidence, prepare typed judgments, and verify claims through MCP tools while leaving editing and execution to the host. | |
| eds-mcp-server edwson/eds-mcp Enables AI agents to access design system tokens and component contracts through MCP, reducing token usage and ensuring consistency. | |
| gemini-web-mcp edwyn211/gemini-web-mcp An MCP server for automated interaction with Gemini's web interface using Playwright and LangGraph, enabling task execution, deep research, file uploads, chat management, and dynamic selector verification via MCP tools. | |
| ClockifyMCP edwyreed/clockify-mcp Enables personal Clockify time tracking through ChatGPT and other MCP clients, allowing users to manage timers, entries, and reports via natural language. | |
| datadog-aws-mcp edxmorales/datadog-aws-mcp A read-only MCP server that exposes tools to query Datadog monitors and logs, and AWS CloudWatch Logs, and to group recurring errors by fingerprint. Designed for use with Claude Code to diagnose issues and propose fixes. | |
| cuporacle-mcp edycutjong/cuporacle-mcp MCP server for FIFA World Cup 2026 data including fixtures, live scores, odds, and knockout bracket, plus a tool (wc_edge) that autonomously pays for premium data via Injective's x402 protocol. | |
| flashframe edycutjong/flashframe Enables automated photosensitivity safety pre-checks on video, resolving flash violations to exact frame spans and using AI to identify the on-screen cause. | |
| Innkeeper edycutjong/innkeeper An MCP server that automates night audit for small hotels by reconciling three ledgers (PMS, card processor, OTA) using AI, adjudicating mismatches with evidence, and cryptographically signing the books. | |
| LineLock edycutjong/linelock A pay-per-pick World Cup edge API on Injective EVM where the x402 USDC receipt IS the pre-kickoff timestamp, plus a free public ledger that CLV-scores every settled pick — losses included. | |
| Unsay edycutjong/unsay An MCP server whose versioned resources can be revised mid-sentence, letting an assistant stop, retract what it just said, and report what changed, who changed it, and how long ago. Every revision is signed, attributable, and hash-chained s | |
| proxmox-mcp edymol/proxmox-mcp A read-only MCP server for Proxmox VE that provides AI assistants with structured visibility into cluster nodes, guests, storage, and Docker workloads. It is designed to prevent any mutating operations by construction. | |
| simple-data-comparator-mcp ee-will-i/simple-data-comparator-mcp MCP server for comparing JSON, XML, XLSX, CSV, and text data using expected templates, returning pass/fail with difference details. | |
| Next.js MCP Server Template ee380/model-context-protocol-mcp-with-next-js A template for deploying Model Context Protocol servers as Next.js routes using the mcp-handler adapter, enabling easy integration of MCP tools, prompts, and resources into any Next.js application. | |
| EEA GeoNetwork MCP Server eea/eea-geonetwork-mcp Enables interaction with the European Environment Agency (EEA) GeoNetwork Catalogue API to search, manage, and edit geographic metadata records. It supports advanced spatial searches, metadata exports in multiple formats, and authenticated | |
| pdf-mcp-server eeager/pdf-mcp-server MCP server for parsing PDF files, enabling Cursor IDE to extract text, metadata, and validate PDFs through Docker. | |
| mysql-mcp-server eeantoss/mysql-mcp-server Enables executing SQL queries, managing databases, and switching between multiple project environments via MCP, without requiring a local MySQL client. |
47,101–47,150 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.