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 |
|---|---|
| data-gov-il-mcp carlc20482/data-gov-il-mcp Enables MCP clients to search and query Israeli government open data through the official CKAN API, covering datasets, resources, tags, organizations, and tabular datastore records. Adds catalog-backed fuzzy discovery with Hebrew normalizat | |
| ReActMCP Web Search carlcampos39/reactmcp Enables AI assistants to perform basic and advanced real-time web searches through the Exa API, returning markdown-formatted results with titles, URLs, publication dates, and content summaries. | |
| Mellos Mapping carlchristou1427/mellos-mapping Enables coding agents to declare an entire system design up front as ghost nodes, then light each node up as it is built and test-verified, exposing a live layered dependency graph in a terminal pane beside the conversation. Structural corr | |
| downloader-mcp carldog/downloader-mcp Enables interaction with SABnzbd (usenet) and qBittorrent (torrent) download clients through their APIs. Each client is optional and only tools for configured clients are registered. | |
| plex-mcp carldog/plex-mcp Lets MCP clients browse and search your Plex media libraries. | |
| servarr-mcp carldog/servarr-mcp An MCP server for the Servarr stack — Sonarr, Radarr, Lidarr, Readarr, and Prowlarr — packaged as a Docker container. Lets an MCP client (Claude Desktop, etc.) browse and search whichever \*arr apps you're running. | |
| foodmcp carleshub/foodmcp MCP server for cooking and recipe management powered by Spoonacular API. Enables natural-language recipe search, ingredient-based discovery, nutrition breakdowns, and ingredient substitutions. | |
| claude-context-server carlformigoni/claude-context-server An MCP server that reads all your Claude Code project memory files and exposes them as tools. Lets any Claude instance — in any project, or via Claude.ai — query your full project history and preferences. | |
| MCP GitHub Project Manager carljeong43618/mcp-github-project-manager Enables AI coding agents and teams to turn GitHub Projects v2 into an agentic task substrate where agents register, claim work atomically, heartbeat, review each other, and track token budgets. It also provides AI-powered PRD generation, ta | |
| mcp-golang-wasm-example carljmosca/mcp-golang-wasm-example An MCP server running in the browser via WebAssembly, providing tools like getCurrentTime and add for streamable HTTP communication. | |
| SpotifyMCP carllapierre/spotify-mcp Enables creating and searching Spotify playlists through natural language using the Model Context Protocol. | |
| MCP Prompt Manager carllee1983/mcp-prompt-manager Git-driven MCP server that manages and provides prompt templates with Handlebars support, enabling teams to collaborate on and version-control reusable, dynamic prompts across AI editors like Cursor and Claude Desktop. | |
| image-recognition-mcp carlleilzj/image-recognition-mcp An MCP server that provides local image recognition on macOS, including OCR, image classification, comprehensive image analysis, and screenshot recognition, all via Apple's Vision framework without any network requests. | |
| whoop-ai-mcp carllombardi51625/whoop-mcp Enables Claude, Codex, and GitHub Copilot to ask natural-language questions about read-only WHOOP recovery, sleep, HRV, strain, workouts, personal baselines, trends, sleep debt, weekly reviews, and HYROX/triathlon tracking. | |
| Keploy carlmax6632-lab/aboutme Keploy is an open-source AI-powered testing platform that helps developers achieve up to 90% test coverage in minutes without writing manual tests. It captures real API traffic and automatically converts it into test cases with mocks and st | |
| Rohlik MCP Server carlmercier84810/rohlik-mcp Enables AI assistants to search grocery products, manage shopping carts, and access account, order, delivery, discount, and personalized meal-suggestion information across multiple European online grocery platforms. | |
| running-coach-mcp carlmichaelkmedieinst/running-coach-mcp Connects AI clients to a personal Intervals.icu account to retrieve recent running activities as clean, normalized JSON. It supports read-only lookups filtered by number of runs and how many days back to search. | |
| deep-code-reasoning carlmoon75/deep-code-reasoning-mcp Enables Claude Code and other MCP clients to escalate complex code investigations to Google's Gemini AI for large-context trace analysis, cross-service impact assessment, performance bottleneck detection, hypothesis testing, and multi-turn | |
| mcp-shodan carlo226/mcp-shodan Gives AI agents read-only access to the Shodan API and Shodan CVEDB, letting them look up open ports, banners, TLS certificates, cloud and location details for IPs, run paged device searches with facet breakdowns, resolve domains and DNS re | |
| mcp-virustotal carlo38525/mcp-virustotal Enables AI agents to analyze files, URLs, IPs, and domains through the VirusTotal v3 API, returning detection reports alongside relationship pivots such as contacted hosts, dropped files, resolutions, and threat actors. It also supports cor | |
| loci-mcp carlo5442/loci Enables AI agents and MCP hosts to search and query a local hybrid vector/BM25 index of scattered notes, docs, PDFs, Office files, web pages, org files, and chat exports. It returns section-level citations so agents can ground answers in th | |
| @cyanheads/pubmed-mcp-server carlo9094/pubmed-mcp-server Enables searching PubMed and Europe PMC, fetching article metadata and full text (via PMC, Europe PMC, or Unpaywall), resolving citations and identifiers, formatting references, and grounding queries in MeSH vocabulary. Runs over STDIO or S | |
| WhatsApp Channel carlo9805/whatsapp-claude-plugin Enables driving an AI coding session directly from WhatsApp chat — bidirectional text, media and voice-note messaging with local transcription, pairing-code/allowlist access control, per-group personalities, and tool-permission approval by | |
| JIRA MCP Server carlodestefano90/jiraagentmcp Enables Claude Code to interact with JIRA through natural language, supporting issue creation, updates, searches, and workflow transitions, designed for enterprise intranets. | |
| contextos-memory carlogiovannifalocco-byte/contextos-v1 Enables AI coding agents to share a local-first, versioned memory of decisions, conventions, tasks, conflicts, and handoffs over MCP and REST. | |
| mcp-home carlogiovannifalocco-byte/mcp-home Enables Alexa+ and other MCP clients to manage a household's shared shopping list, bills, chores, and pantry expiry through natural-language voice interactions. It provides read-only briefings and confirmation-gated updates for recurring ho | |
| Mind Keg MCP carloluisito/mindkeg-mcp A persistent memory server that stores and retrieves atomic coding insights like architectural decisions and debugging patterns for AI agents. It enables agents to maintain institutional knowledge across sessions using semantic search and l | |
| git-context-mcp carlomeazza/git-context-mcp A secure MCP server for local Git operations with path traversal protection, input validation, and 15 supported Git commands. | |
| swissdevjobs carlomendoza552/swissdevjobs-cli Enables AI agents to search, filter by salary, and apply to ~4,700 salary-transparent tech jobs across nine boards and eight countries, while remembering past applications so duplicates are hidden and returning results as JSON-first data. | |
| Ombre Brain carlomitsotakis2043/ombre-brain Enables Claude to maintain persistent cross-conversation emotional memories tagged by valence and arousal, stored as Obsidian-compatible Markdown, with natural decay, active surfacing, merging, and retrieval through MCP tools. | |
| mcp-search carlonicolini/mcp-search Enables advanced reasoning for large language models through Monte Carlo Tree Search and Beam Search, allowing exploration of multiple reasoning paths and improved decision-making. | |
| JMComic AI MCP Server carlophillips26713/jmcomic-ai Enables AI clients to search, browse, favorite, download, and post-process comics and chapters through standardized tools and resources, with task control, configuration updates, and account login. | |
| Hydra Vision MCP carlory/hydra-vision-mcp Connects a text-only coding agent to an OpenAI-compatible vision model so it can analyze images, screenshots, and UI mockups. Provides tools for general image understanding, OCR and code transcription, error-screenshot diagnosis, technical | |
| RobotMCP carlos-eduardo-1984/robotframeworkmcp Enables AI agents to discover Robot Framework keywords, execute test steps live against web, mobile, API, database, or desktop targets, and generate clean .robot test suites from plain-English instructions. | |
| rf-mcp carlos-eduardo-1984/robotmcp_poc Enables AI agents to drive Robot Framework through natural language, discovering keywords, running live test steps across web, mobile, API, database, and desktop targets, and generating clean .robot test suites from plain-English instructio | |
| juris-oficial carlos-jus/juris-oficial Enables searching Brazilian court jurisprudence directly in official sources (STF, STJ, TNU, TRF1–6, Turmas Recursais, TRUs, and BNP/CNJ qualified precedents), returning literal ementas, panel, rapporteur, dates, and full-text links. It wor | |
| Siigo MCP carlos-ortiz23/siigo-mcp Remote MCP server that connects the Siigo Colombia API to any standard MCP client, exposing 28 tools for managing customers, products, invoices, purchases, quotations, receipts, accounting, and business analytics while keeping Siigo credent | |
| NotebookLM MCP Server carlos1271/notebooklm-mcp Enables agents to query Google NotebookLM notebooks with citation-backed answers and to generate and download Studio artifacts such as audio overviews, video, infographics, reports, presentations, data tables, flashcards, quizzes and mind m | |
| mcp-shodan carlos7164/mcp-shodan Enables AI agents to query the Shodan API and Shodan CVEDB for internet-connected device reconnaissance — open ports, banners, TLS certificates, DNS records, and CVE/CPE vulnerability intelligence with CVSS, EPSS, and KEV status. It exposes | |
| featuriq carlosalvite/featuriq-mcp Connect your AI assistant to Featuriq — the product feedback and roadmap tool for SaaS teams.
Browse top feature requests, search feedback with natural language, update statuses, notify users when features ship, and manage your roadmap — a | |
| MCP Dataverse Server carlosarmenta75/dataverse_mcp_server Enables CRUD operations and schema exploration on Microsoft Dataverse databases using service principal authentication. It allows users to query records with OData filters, manage table entries, and retrieve metadata through a standardized | |
| Read-only Analytics MCP Server carlosatom8/readonly-analytics-mcp-server Enables analytics agents to run validated read-only SQL against a warehouse API with enforced row limits, required JSON responses, and secret-safe audit metadata. | |
| MCP Server CCXT carlosatta/mcp-server-ccxt Provides access to cryptocurrency exchange APIs through CCXT, enabling market data retrieval, account balance checks, and trading operations across multiple exchanges like Binance, Coinbase, and Kraken with built-in security features. | |
| Activity Reporting MCP Server carlosazaustre/advocu-mcp-server Enables Google Developer Experts to report various activities (content creation, speaking engagements, workshops, mentoring) through AI conversational interfaces by connecting Advocu API with Model Context Protocol. | |
| fcvm carlosbravoa/fcvm Disposable Firecracker microVMs as sandboxes for AI agents: run commands and move files in any OCI image, snapshot and fork running copies in ~150 ms, egress allowlists. Self-hosted; needs Linux with KVM. | |
| Cinema4D MCP Server carlosbruno857/cinema4d-mcp Connects Claude to Cinema 4D so users can build and manipulate 3D scenes through natural-language prompts, including adding primitives, cameras, lights, materials, MoGraph cloners and dynamics, executing custom Python, and rendering preview | |
| ScrapeGraph MCP Server carlosc34/scrapegraph-mcp Enables language models to scrape and extract structured data from web pages, run searches, manage multi-page crawls, generate schemas, and monitor scheduled jobs through an AI-powered web scraping API. | |
| swissdevjobs carlosc6590/swissdevjobs-cli Enables AI agents to search ~4,700 salary-transparent tech jobs across nine boards and eight countries, filter by salary, tech, remote work and seniority, and apply — while remembering which postings were already applied to so they never re | |
| Camino MCP carlosceja27/camino-mcp Enables local AI clients to query Santa Clara University's Camino courses, assignments, upcoming/overdue work, and visible grade summaries through a read-only MCP server. | |
| Ombre Brain carloschoi262/ombre-brain Gives Claude persistent cross-conversation memory by tagging each entry with Russell's valence/arousal coordinates, storing them as Obsidian-compatible Markdown, and letting emotionally intense or unresolved memories decay slowly and surfac |
36,851–36,900 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.