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 |
|---|---|
| apple2-mcp benj-edwards/apple2-mcp An MCP server that wraps the Bobbin Apple II emulator, providing tools for memory access, screen reading, BASIC program execution, and disk image management to enable autonomous experimentation with Apple II control techniques. | |
| deliver-cli benjamesmurray/deliver-cli Transforms AI agents into spec-driven product engineers by managing the software project lifecycle through requirements, design, implementation, and archiving phases with state-aware MCP tools. | |
| Bruno MCP Server benjamiinn1/bruno-build-mcp Automatically generates and maintains Bruno API test collections for CRUD JSON APIs. Developers describe endpoint changes and the server creates ready-to-run test suites with environment-driven configuration. | |
| google-maps-mcp benjamin-di/google-maps-mcp Enables AI assistants to look up travel times for driving, walking and cycling, turn-by-turn directions, multi-stop route plans, place details with opening hours, and map images, all from Google's map data. It works read-only without an API | |
| photos-mcp benjamin-di/google-photos-mcp Enables an assistant to find and view photos in one Google Photos library using Google Photos' own search and image views, with no write access. | |
| hktvmall-mcp benjamin-di/hktvmall-mcp-server Enables an AI assistant to search a Hong Kong online mall for a whole shopping list, compare prices and unit prices, show product cards with photos, read order history, and set quantities in the user's real basket. It never places, pays for | |
| Newsclip MCP benjamin-di/newsclip-mcp Enables searching, reading, and privately archiving news from supported publisher subscriptions, with per-user authentication via human sign-in links and separate tokens, sessions, and archives. | |
| uber-mcp benjamin-di/uber-mcp-server Enables an AI assistant to find places, compare live Uber ride prices, request rides behind a price-capped preview/confirm step, track or cancel active rides, and review trip history. | |
| Wiki Hub MCP benjamin-di/wiki-hub-server Provides OAuth-secured MCP tools for working with private, Git-backed Markdown knowledge bases: searching pages, reading content, ingesting new source documents, retrieving cited original files, and reviewing history or rolling back accepte | |
| wezas-africa-toolkit benjamin09e/wezas-toolkit Enables AI assistants and Python applications to validate African phone numbers, convert CFA franc currencies with fixed parity, and retrieve country details such as dialing codes, time zones, and legal holidays, all offline without API key | |
| Email MCP Server benjamin45370/email-mcp Enables AI assistants to read, search, send, reply, forward, organize, schedule, and analyze email across multiple IMAP/SMTP accounts, with drafts, provider-aware labels, calendar extraction, verification-code catching, and a real-time inbo | |
| FlowLens MCP benjamin45370/flowlens-mcp-server Enables coding agents to access recorded browser flows with user actions, network, console, storage, DOM events, and screen recording so they can debug, report bugs, perform regression testing, and generate Playwright tests without reproduc | |
| ScrapeGraph MCP Server benjamin4774/scrapegraph-mcp Enables language models and AI assistants to perform AI-powered web scraping, extraction, search, crawling, schema generation, and monitor management through the ScrapeGraph API. | |
| Drug Interaction MCP Server benjamin920101/drug-interaction-mcp Enables querying of Chinese-Western medicine interactions with comprehensive drug information, risk assessment, and clinical recommendations. Supports searching medicines, checking interactions individually or in batches, and provides safet | |
| Taiwan FDA Drug Search MCP Server benjamin920101/fda-taiwan-mcp Enables searching Taiwan FDA drug database by various criteria (English/Chinese names, license numbers, ingredients), retrieving drug information, and downloading PDF package inserts with CAPTCHA verification support. | |
| send-to-kindle-mcp benjaminbini/send-to-kindle-mcp Sends document files (EPUB, PDF, DOCX, etc.) to your Kindle via email using Amazon's Send to Kindle feature. | |
| Proxmox MCP Server benjaminduthe/proxmox-mcp Enables Claude to manage Proxmox VE infrastructure — VMs, LXC containers, snapshots, storage, and more. | |
| Google Drive MCP Server benjamine/gdrive-mcp Enables searching, reading, and commenting on Google Drive documents through secure OAuth2 authentication, with support for creating styled notes directly in document content. | |
| diff-mcp benjamine/jsondiffpatch compare 2 text or data (supports text diffs, json, json5, yaml, toml, xml and html). powered by jsondiffpatch | |
| Office Live MCP benjaminfrombe/office-live-mcp Enables AI agents to read and edit live, already-open Word, Excel and PowerPoint documents rather than files on disk, with formulas calculated and changes visible on screen as they happen. Exposes 85 tools across the three apps plus raw Off | |
| bizagi-modeler-local benjaminjaldin/bizagi-contexto-procesos Enables MCP clients to connect to saved Bizagi Modeler .bpm files, detect the local installation, open models, and retrieve diagram entities and connections with refresh after saves, all within a user-authorized folder. | |
| Dart LSP MCP benjaminkobjolke/dart-lsp-mcp A Model Context Protocol (MCP) server that provides Dart language analysis capabilities to Claude Code, enabling real-time diagnostics, code navigation, and symbol search for Dart projects. | |
| BOSS 直聘 MCP Server benjaminmorales256/boss-zhipin-mcp Enables AI assistants to automate recruiter-side hiring on the BOSS Zhipin platform — batch-searching candidates across multiple keywords with deduplication, screenshotting and OCR-ing Canvas-rendered resumes, extracting permanent share lin | |
| caldav-mcp benjaminmoreira67112/caldav-mcp Enables AI assistants to connect to CalDAV servers and manage calendars, events, and tasks through MCP tools. | |
| Sendiment benjaminmower/sendiment cast a thought into the river of consciousness | |
| Sugar benjaminn581/sugar Enables AI coding agents to read and write persistent, local-first memory across sessions and projects, with semantic search, global guidelines, and optional autonomous task execution. | |
| workbuddy-subagent-bridge benjaminnh/workbuddy-subagent-bridge Enables a main agent to delegate execution tasks to WorkBuddy through MCP, supporting multi-turn sessions, task status tracking, permission approvals, execution reports, and CLI fallback. | |
| chrome-devtools-mcp benjaminr/chrome-devtools-mcp An MCP Server for Chrome DevTools, following the Chrome DevTools Protocol. Integrates with Claude Desktop and Claude Code. | |
| mcp-koii benjaminr/mcp-koii Enables controlling a Teenage Engineering EP-133 K.O. II sampler via MIDI through natural language, allowing note playback, pattern creation, and drum pattern sequencing. | |
| easyhook-mcp-server benjaminrm10/easyhook-mcp-server Allows sending WhatsApp messages via Easyhook from MCP clients like Codex and Claude, with a fixed sender and allowlisted contacts for security. | |
| nexusgrid-mcp benjaminsamson0210/nexusgrid-mcp Enables AI coding agents and LLMs to scrape web pages into clean Markdown, audit email deliverability and authentication records, and verify email signup fraud risk through MCP tools. | |
| Claude Telegram Supercharged benjaminwagner657/claude-telegram-supercharged Enables running Claude Code around the clock through Telegram, with voice notes exchanged in both directions, a self-healing daemon that keeps sessions alive, memory and history preserved across restarts, and a live view of every tool call | |
| browser-bridge benjaminwood/arc-playwright-bridge Enables an AI agent on another machine to drive a real, logged-in Arc profile on a Mac over SSH, attaching Playwright to that profile's existing tabs, opening new ones in a chosen sidebar space, and skipping configured hidden origins. | |
| Remote MCP Server on Cloudflare benjamitm/remote-mcp-server-prod A template for deploying MCP servers on Cloudflare Workers with OAuth authentication, enabling remote access to MCP tools via SSE transport from Claude Desktop and other MCP clients. | |
| Global WebMCP benjang032/globalwebmcp Enables teaching an agent a website once and then calling its versioned actions like functions through MCP tools, with strict validation and safe execution of public site operations. | |
| palace-tools benjhardy/palace-agent-tools Exposes AWS Palace finite-element electromagnetic simulation as a set of JSON-in/JSON-out tools, so an agent can build geometry and meshes, generate and lint configs, estimate cost, guard and launch runs, and read back S-parameters, eigenmo | |
| QualityPilot benjik50/qualitypilot-ai Enables querying quality procedures in natural language, retrieving relevant passages via RAG, and calculating supplier KPIs such as PPM and non-conformity rates through MCP-exposed tools. | |
| github-mcp-lite benjimkh/github-mcp-lite A lightweight MCP server for GitHub with nine essential tools for reading files, managing pull requests, and creating issues, using personal access token authentication. | |
| claude-fde-showcase benjiscollector/claude-fde-showcase MCP server demonstrating FDE primitives: TF-IDF search, extractive summarisation, and an AST-sandboxed calculator. Designed for offline, deterministic tool use. | |
| mcp-arabic-toolkit benjiscollector/mcp-arabic-toolkit MCP server exposing Arabic text utilities: normalisation, tashkeel stripping, transliteration, heuristic dialect detection, and token counting. | |
| French Law MCP Server benjonx/legal Enables querying and retrieving legal texts and decisions from French public APIs Légifrance and JudiLibre for legal research. | |
| saas.tf MCP Server benjp009/saas-tf-mcp-server Enables users to get free subdomains with automatic DNS provisioning directly through LLM-powered tools like Claude Code. | |
| databricks-sql-mcp benkeeley9/databricks-sql-mcp Enables AI assistants to execute SQL queries and explore databases, tables, and catalogs on Databricks using Unity Catalog. | |
| AI-to-CATIA MCP benkeith1989-star/ai-to-catia-mcp Enables safe, AI-driven CATIA V5 modeling through 36 guarded tools for creating and inspecting parametric CATParts, including sketch, part design, GSD/surface, and transaction operations, while preventing mutation of existing documents. | |
| mcp-memories benkipnis/mcp-memories Captures code build sessions and publishes them to a shared team knowledge base, making past work discoverable via semantic search. | |
| Virtual Engineer MCP Server benkipnis/virtual_engineer Enables AI-assisted chiller troubleshooting by exposing MCP tools to retrieve operational data, alarms, telemetry, and service history from MongoDB Atlas, plus vector and hybrid search over knowledge bases. | |
| Day One MCP Server benlen10/dayone-mcp Read-only MCP server for Day One journal on macOS, enabling search and browsing of entries, photos, and metadata. | |
| Masset MCP Tools benmasset/masset-mcp-tools Provides honest A/B test verdicts using interactive cards, with statistical calculations for conversion rates, sample sizes, and Bayesian probability-to-beat. | |
| job-scrapy benmishal/job-scrapy Enables real-time job discovery across multiple job boards and remote APIs, providing normalized and deduplicated job listings to AI agents via MCP. | |
| mcp-tldr benmousaamine/mcp-tldr MCP server that exposes llm-tldr code analysis tools (tree, structure, context, search, impact, etc.) to MCP clients like Claude Code for project understanding and debugging. |
33,601–33,650 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.