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 |
|---|---|
| openalex-mcp cguo19/openalex-mcp Provides access to the OpenAlex API, enabling search and retrieval of scholarly works, authors, sources, and more. | |
| BigQuery MCP server cguo63884/mcp-server-bigquery Enables LLMs to inspect database schemas and execute SQL queries, list tables, and describe table schemas. | |
| io.github.cdnsteve/sugar cguo63884/sugar Provides a local-first persistent memory layer that lets AI coding agents store, search, and recall project-specific and cross-project knowledge across sessions. It also offers optional autonomous task queue execution, built on the same mem | |
| Gmail AutoAuth MCP Server cgutiguti/gmail-mcp-server Enables AI assistants to manage Gmail through natural language interactions with auto-authentication, including sending, reading, searching, and organizing emails and labels. | |
| MCP Server Demo ch-122/mcp-server A monorepo containing three Model Context Protocol (MCP) servers that enable natural language interaction with databases, GitHub repositories, and time/timezone services. | |
| Linear Cache MCP ch-raph/linear-cache-mcp Cache-first MCP server for Linear that provides tools to search, read, create, update, and comment on issues and projects, caching data locally to reduce API calls and respect rate limits. | |
| Multilogin Browser MCP ch-yousafali/mcp-server Enables AI agents to control Multilogin X anti-detect browser profiles via a streamable HTTP MCP interface. Supports profile management, session handling, and a broad catalog of browser actions such as navigation, clicking, typing, screensh | |
| Claude PM MCP ch1kim0n1/claude-pm-mcp Turns Claude Code into an engineering project manager by orchestrating feature development with isolated Git worktrees, structured task validation, and approval-gated integration. | |
| Pentest Tools MCP Server ch1nhpd/pentest-tools-mcp-server An MCP server that integrates various penetration testing tools, enabling security professionals to perform reconnaissance, vulnerability scanning, and API testing through natural language commands in compatible LLM clients like Claude Desk | |
| notification-mcp ch1y1z1/notification-mcp Enables sending push notifications via Bark and PushDeer services, supporting message, markdown, and image notifications. | |
| cve-reference-mcp ch201y00n/cve-mcp Collects and returns reference materials for CVE write-ups from sources like NVD, CIRCL, OSV, EPSS, and GitHub Advisory. | |
| Codex Model Router ch3sh-lc/codex-model-router Enables users to route Codex desktop chats to subscription models or their own OpenAI-compatible providers by adding Base URLs, API keys, and model catalogs from an in-app settings panel. Keeps keys DPAPI-encrypted, supports Responses and C | |
| Thread Contract MCP Server ch998244353/thread-contract-mcp Thread Contract is a local runtime contract layer for AI coding-agent threads. It lets users pin explicit, thread-scoped rules without turning them into project policy or long-term memory. | |
| cloudflare-playwright-mcp cha0s-repo/playwright-mcp-example MCP server that enables AI assistants to control a browser through Playwright, supporting tasks like navigation, typing, clicking, and taking screenshots. | |
| umodel-mcp cha1007/umodel-mcp MCP server that enables natural language listing, unpacking, and exporting of Unreal Engine game resources via umodel, including AES-encrypted pak support. | |
| TossInvest MCP cha2hyun/tossinvest-mcp Enables querying Korean and US stock information, quotes, account assets, and order history via Toss Securities Open API. Optionally supports order placement with preview and confirmation safeguards. | |
| Stepgate chaarangan/stepgate Enables agents to run gated, API-only YAML stepfiles in order on the client's own model, making each step's output satisfy schemas, predicates, or HTTP verifiers before proceeding. It also handles API calls, credentials, retries, and hash-c | |
| mcp-stm-montevideo chaba11/mcp-stm-montevideo MCP server exposing Montevideo public transportation data (STM) as tools for AI assistants, enabling natural language queries about routes, stops, arrivals, and trip planning. | |
| toss-mcp chabinhwang/toss-mcp Provides AI agents with access to Toss developer documentation and icon catalogs, enabling search and retrieval of official docs and icon usage recommendations. | |
| MCP Icon Generator chacebot/mcp-icon-generator Generates app icons with customizable text, colors, and sizes, including complete iOS app icon sets. | |
| MCP Xcode Server chacebot/mcp-xcode-server An MCP server that provides tools for programmatically interacting with Xcode projects, enabling users to manage targets, schemes, and files. It supports essential development tasks including building, testing, and cleaning projects through | |
| mcp-lightrag chacharin/mcp-lightrag Enables AI agents to connect to a LightRAG knowledge-graph server over HTTP, exposing 33 tools for querying, managing documents, and editing knowledge graph entities and relations with proper authentication and error handling. | |
| Perplexity MCP Server chacharin/mcp-perplexity Enables MCP-compatible AI agents to search the web and obtain cited answers, step-by-step reasoning, and deep research through hosted APIs. | |
| mcp-thai-proofread chacharin/mcp-thai-proof-md Enables AI agents to proofread and correct Thai spelling in Markdown files or plain text through any OpenAI-compatible LLM endpoint, fixing spelling only without rewriting content. It validates each LLM response line-by-line to protect numb | |
| mulchd chachegs/mulchd Self-hosted MCP server for storing and serving structured team knowledge, enabling AI sessions to load relevant team context without re-prompting. | |
| elsevier-mcp chackchang/elsevier_mcp MCP server for Elsevier Developer APIs, providing access to Scopus, ScienceDirect, and other academic databases with tools for search and retrieval. | |
| Railway OpenWebUI MCP Server chad-atexpedient/railway-openwebui-tool Enables deploying, managing, and monitoring OpenWebUI instances on the Railway platform via natural language commands. | |
| spotdraft-mcp chad-hohn-radai/spotdraft-mcp MCP server that gives Claude safety-gated access to SpotDraft contracts, enabling querying and managing contracts, templates, counterparties, obligations, and analytics through natural language. | |
| Warden chadcorp/warden Warden provides a trusted skill brain for open agents, enabling secure execution of curated, cryptographically-signed skills with sandboxing, trust scoring, and transparency logging, all with zero third-party dependencies. | |
| eufy-lock-codes chaddbrenner/eufy-lock-codes A local MCP server for managing Eufy smart-lock access codes across rental properties, with safety features like dry-run planning and explicit confirmation for write operations. | |
| Remote MCP Server Authless chadismoving/remote-mcp-server-authless Enables deployment of MCP servers to Cloudflare Workers without authentication requirements. Allows users to create custom tools and connect them to MCP clients like Claude Desktop or Cloudflare AI Playground through remote URLs. | |
| NetBox MCP Server chadkunsman/netbox_mcp A read-only FastMCP server that enables AI assistants to query and retrieve network infrastructure information from NetBox using natural language. | |
| Zoom MCP Server chadkunsman/zoom-mcp Enables intelligent monitoring and management of Zoom rooms across multiple sites using smart location resolution and hierarchical discovery. It provides tools for checking room status, retrieving location hierarchies, and accessing detaile | |
| anki-mcp chadlis/anki-mcp Enables reading and writing to your local Anki collection through AnkiConnect, allowing creation, update, and search of notes and decks from MCP clients like Claude Desktop. | |
| ynab-mcp chadlis/ynab-mcp Enables reading and writing YNAB budget data, such as listing budgets, accounts, categories, transactions, and creating or updating transactions, through natural language commands. | |
| sleeper-caffeine MCP chadnickbok/sleeper-caffeine Provides read-only access to Sleeper fantasy football leagues, enabling team snapshots, available players, matchups, trade context, and league history through standardized MCP tools. | |
| Deepgram Agentic Tools MCP Server chadpapineau/deepgram-mcp-gateway A self-hosted MCP server that wraps Deepgram's speech-to-text, text-to-speech, text analysis, and usage APIs for use with CyberArk Secure AI Agents (SAIA). | |
| Motion MCP Server chadramos/motion-mcp-server Bridges Motion's API with LLMs via MCP, enabling natural language management of tasks, projects, schedules, and more. | |
| maya-mcp-server chadrik/maya-mcp-server MCP server for interacting with Autodesk Maya sessions, enabling multi-session management, arbitrary Python execution, and streaming output capture. | |
| assessor-lookup-mcp chadru/assessor-lookup-public Enables AI agents to look up county assessor public records for properties, check MLS discrepancies against public data, and discover new county assessor sources, all via a local MCP server for real-estate appraisal workflows. | |
| Ruddur MCP Server chadsly/ruddur-mcp An MCP server that enables natural language interaction with the Ruddur creator API, supporting content management tasks like creating headings and waypoints, with OAuth for hosted deployments and token-based local stdio mode. | |
| upnote-mcp chadthornton/upnote-mcp Enables AI assistants to interact with Upnote via its x-callback-url API, allowing creation of notes, notebooks, tag management, and search. It also supports navigation to various Upnote sections and custom filters. | |
| worksuite-mcp chadvdwww-hash/worksuite-mcp Connects AI assistants to Google Workspace (Gmail, Drive, Docs, Sheets, Slides, Calendar, Tasks, Forms, Chat, Contacts) through your own OAuth app, providing around 120 tools for managing email, files, calendar, and more. | |
| shared-agent-memory-mcp chaerulcp/shared-agent-memory-mcp Enables AI coding agents to store and retrieve durable project decisions, conventions, environment notes, and bug fixes in Notion, with optional Obsidian Vault dual-write. | |
| Tomato Writer MCP chaeunwooy/tomato-writer-mcp Enables AI clients to interact with the Fanqie free novel author backend, including listing works, fetching chapter statuses, retrieving statistics, and publishing or drafting chapters. | |
| 네이버 길찾기 MCP 서버 chaeyun06/naver-directions-mcp Enables users to access Naver Maps API functionality including directions, place search, geocoding, and reverse geocoding through a Model Context Protocol server. | |
| smsf-rules-mcp chaglar/smsf-rules-mcp Enables users to look up and search Australian SMSF rules — contribution caps, total super balance and Division 296 thresholds, pension minimums and lodgment deadlines — and to compute contribution headroom, every answer carrying an ato.gov | |
| MachineBridge Unified Server chahakshahcs5/machinebridhe-js Enables AI clients and agents to execute commands in real interactive PTY sessions, manage host filesystem operations, and access 15 MCP tools via Stdio, SSE, or REST with secure API-key authentication. | |
| saavn-mcp chahalhasanpreetsingh-prog/saavn-mcp Enables an AI agent to download songs from plain-language queries or JioSaavn links and add them to a self-hosted Navidrome library with tags, cover art, and automatic rescan. | |
| jira-mcp-server chaholl/jira-mcp-server An MCP server that lets AI assistants create, search, and manage Jira issues using Chrome browser session cookies for authentication, eliminating the need for API tokens. |
37,751–37,800 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.