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 |
|---|---|
| Local MCP Dev Runner cosmgrid/local-mcp-dev-runner Enables AI assistants to securely inspect local Git repositories and perform restricted write operations in isolated worktrees, with sandboxed script execution and limited GitHub repository management. | |
| matrix-mcp cosmic-fire-eng/matrix-mcp MCP server for Matrix that lets Claude list rooms, search/read messages, send messages and files, react, create rooms, and invite users, with multi-homeserver support and safe-by-default writes; no end-to-end encryption. | |
| local-brain-mcp cosmiccoder200x-sys/local-brain-mcp Enables AI coding assistants to recall past lessons, bugs, and conventions from local git history with sub-5ms latency and full privacy. | |
| Windows Admin MCP cosmicjedi/windows-admin-mcp Enables remote Windows server administration and troubleshooting via WinRM and SSH PowerShell protocols. | |
| Blender Antigravity MCP cosmictraveler002/blender_antigravity_mcp Enables AI assistants in Google Antigravity to drive Blender for 2D-image-to-3D reconstruction, running a multi-stage pipeline to analyze reference images, generate models, compare renders, refine geometry, and produce reports. | |
| mail-mcp-custom cosmingherghe/mail-mcp-custom Custom MCP server for Gmail integration, enabling email listing, reading, searching, and sending via natural language. | |
| Confluence MCP cosmix/confluence-mcp A Model Context Protocol server that enables AI assistants to interact with Confluence content, supporting operations like retrieving, searching, creating, and updating pages and spaces. | |
| JIRA MCP Server cosmix/jira-mcp Provides an interface to access and manage JIRA data through the Model Context Protocol, offering features like relationship tracking, data cleaning, and contextual insights for AI applications. | |
| Linear MCP Server cosmix/linear-mcp Provides a Model Context Protocol interface for accessing Linear's issue tracking system, enabling users to query and search issues with TypeScript type safety and robust error handling. | |
| MCP Web Docs cosmocoder/mcp-web-docs A self-hosted MCP server that crawls, indexes, and searches documentation from any website locally, including private sites requiring authentication. It provides hybrid search capabilities and local embedding generation to maintain privacy | |
| JacBrain cosmonautjones/jacbrain Enables AI coding agents to store and retrieve versioned project knowledge and compiler-verified Jac snippets, so they can reuse relevant context across tasks and validate candidate code with the Jac compiler. | |
| mcp-off-shelf cosmonic-labs/mcp-server-template-ts-p3 A TypeScript MCP server packaged as a self-contained WASI p3 WebAssembly component that exposes basic arithmetic, echo, time, and bounded outbound HTTP tools, plus discovery, health, and embedded skill resources. It supports streaming SSE p | |
| mcp-server-template cosmonic-labs/petstore-mcp-server Template repository for building a WebAssembly component MCP server using wasmCloud. | |
| Cosmos SDK Migration MCP Server cosmos/cosmos-sdk-migration-mcp An MCP server for migrating Cosmos SDK chains from v0.50+ to v0.54. | |
| Hermes Supabase MCP cosmosluna/hermes-supabase-mcp Enables CRUD operations on Supabase structured data through PostgREST tools without exposing arbitrary SQL. Access can be constrained to specific tables via an allowed-tables setting. | |
| Cosmos Pay MCP Server cosmospay/cosmospay-mcp Enables AI agents to create stablecoin payment links with QR codes and hosted checkout pages, and to check, list, validate or delete payments across Stellar, Solana and Monad using an API key. Read-only mode, scoped keys and tool annotation | |
| MUSTer MCP Server cosmostima/muster_mcp Enables LLM interaction with the Macau University of Science and Technology (M.U.S.T.) campus system, including automated login to Wemust and Moodle, retrieving class schedules, checking assignments and deadlines, downloading course materia | |
| vnsh-mcp costa10559/vnsh Enables AI agents to create, read, update, list, and restore end-to-end encrypted workspaces and artifacts that any other agent or person can access through one shareable link, with version-checked writes and read-only or public sharing var | |
| Basecamp MCP Server costa368/basecamp-mcp-server Enables MCP-capable clients to authenticate with Basecamp 3 via OAuth and read or manage projects, todos, messages, card tables, documents, uploads, comments, events, webhooks, and search through its tool set. | |
| U.S. Census Bureau Data API MCP costa5817/us-census-bureau-data-api-mcp Connects AI assistants with official U.S. Census Bureau statistics, enabling them to discover available datasets, resolve place names to FIPS geography codes, and retrieve aggregate demographic and economic data through natural language. It | |
| fcp-mcp-server costa8077/fcp-mcp-server Enables Claude to read, edit, and generate real Final Cut Pro XML timelines — trimming and splitting clips, detecting cuts and flash frames, placing markers from transcripts or beats, assigning roles, building rough cuts, and exporting to o | |
| MCP Todo Server costaadiego1989/mcp-todo-server A CRUD todo list server that exposes tools to create, list, and conclude tasks, compatible with any MCP host. | |
| CosVoice cosvoice/cosvoice-mcp Gives an AI assistant its own real phone number and email address so it can place and answer calls, book reservations by phone, send and receive texts, and read written summaries, structured details, and transcripts of every conversation. A | |
| Swiss Legal MCP Server cotadavidch-ctrl/swiss-legal-mcp-public Enables legal research and procedural analysis from verified Swiss legal sources, with tools for case resolution, legal drafting, and multirole argument evaluation. | |
| Scraper MCP cotdp/scraper-mcp A context-optimized web scraping server that converts HTML to markdown/text and applies CSS selectors server-side, reducing token usage by 70-90% while providing AI tools with clean, filtered web content. | |
| Google Discover MCP Server cotletdev/google-discover-api-mcp-server Enables MCP clients to read the signed-out Google Discover feed as structured JSON without a Google account, including each card's position, title, publisher, URL, domain, type, age, and image. It also lets agents check whether a domain app | |
| Joist MCP Server cotletdev/joist-api-mcp-server Enables AI agents and scripts to access clients, estimates, invoices, payments, and reports from a user's own Joist account through their Chrome session, over MCP or a local REST API. | |
| Omio MCP Server cotletdev/omio-api-mcp-server Enables AI agents to search trains, buses, flights and ferries, compare fares and refund conditions, find the cheapest travel day, and reserve a trip up to a checkout link, plus read existing bookings and saved travelers, all through your o | |
| Quora MCP Server cotletdev/quora-api-mcp-server Enables AI agents and scripts to search and read Quora questions and answers, post answers and questions, handle answer requests and notifications, and work with Spaces, posts, and submissions through 13 tools running on your own signed-in | |
| SoundCloud MCP Server cotletdev/soundcloud-api-mcp-server Runs on a self-hosted Python process using your own signed-in Chrome session, exposing ten tools over stdio, HTTP, REST, and Python so an agent can read your profile and any profile by link, list tracks including private ones and their secr | |
| Maya MCP cotton357/maya_mcp Enables AI assistants like Claude Desktop to control Autodesk Maya via natural language through the Model Context Protocol. | |
| ytt-mcp cottongeeks/ytt-mcp An MCP server designed to fetch transcripts for YouTube videos. It enables AI tools to access video text content for tasks like summarization, analysis, and key takeaway extraction. | |
| dsh-yuque-mcp couchette/dsh-yuque-mcp Lets an agent read and write Yuque knowledge bases, documents, tables of contents, notes, and resources through the Yuque open API. It bundles the upstream server and prompts for a personal token at first run or when a token expires, so all | |
| kjlahsdjkashdjhkasdkajshd couchrishi/sd-for-designers kjlahsdjkashd | |
| mcp-server-requests coucya/mcp-server-requests Web Content Retrieval (full webpage, filtered content, or Markdown-converted), Custom User-Agent, Multi-HTTP Method Support (GET/POST/PUT/DELETE/PATCH), LLM-Controlled Request Headers, LLM-Accessible Response Headers, and more. | |
| Arcane Docker MCP Server cougz/arcane-mcp-server Enables managing Docker environments through Arcane via natural language, providing tools for containers, images, volumes, networks, and Docker Compose stacks. | |
| gemini-imggen couhie/mcp-gemini-imggen Enables image generation and transformation using Gemini API, returning file paths instead of base64 data to avoid token limit issues in Claude Code. | |
| meok-aaif-agent-card-mcp council-of-ai-csoai-org/csoai-aaif-agent-card-mcp Publishes AAIF Agent Cards and bridges A2A and OASF to AAIF, with built-in EU AI Act compliance. | |
| Meok Abci Bridge MCP council-of-ai-csoai-org/csoai-abci-bridge-mcp Read-only Tendermint/Cosmos blockchain query for agents, with 10 pre-configured mainnets and custom RPC support. | |
| MEOK CRA Annex Iv Classifier MCP council-of-ai-csoai-org/csoai-cra-annex-iv-classifier-mcp MEOK CRA Annex Iv Classifier - MCP server providing AI-powered tools and automation by MEOK AI Labs | |
| meok-eu-ai-act-art-13-ifu-mcp council-of-ai-csoai-org/csoai-eu-ai-act-art-13-ifu-mcp Generates EU AI Act Article 13 Instructions for Use (IFU) and provides a FRIA crosswalk, enabling providers to create transparency documentation compliant with the EU AI Act. | |
| meok-eu-ai-act-art-26-fria-mcp council-of-ai-csoai-org/csoai-eu-ai-act-art-26-fria-mcp Generates EU AI Act Article 26(9) FRIA documents with 9-element structure, DPIA crosswalk, and signed attestation for compliance. | |
| MEOK NIS2 De Register MCP council-of-ai-csoai-org/csoai-nis2-de-register-mcp MEOK NIS2 De Register - MCP server providing AI-powered tools and automation by MEOK AI Labs | |
| MEOK Omnibus Tracker MCP council-of-ai-csoai-org/csoai-omnibus-tracker-mcp MEOK Omnibus Tracker - MCP server providing AI-powered tools and automation by MEOK AI Labs | |
| BusTime MCP countaight/bustime-mcp Wraps the CTA Bus Tracker API to provide live access to bus routes, stops, predictions, vehicle locations, patterns, and detours via MCP clients. | |
| Everclear MCP Server counter-measure/everclear-mcp Provides Claude AI with direct access to Everclear.org's API endpoints for querying intents, invoices, analytics, and performing blockchain operations. | |
| mealie-mcp-ts counterbeing/mealie-mcp-ts Enables AI assistants to manage recipes, meal plans, and shopping lists in Mealie through natural language, supporting CRUD operations and URL-based recipe imports. | |
| Browser MCP counterm/browser-mcp Exposes browser tabs to any MCP client through a local broker, letting agents discover pages, take text snapshots with numbered elements, and drive them via click, type, navigate, eval, network/log inspection, and screenshots. Supports stdi | |
| Countersign countersign-network/packages Kill switch + spend guard for AI agents that spend money: one policy, one sub-second freeze, and one signed audit ledger across every wallet vendor at once (Coinbase, Turnkey, Openfort + cards). Tools include a pre-flight request_spend guar | |
| Coupler MCP Server coupler-io/coupler-io-mcp-server Coupler.io MCP server lets AI assistants query business data from 400+ apps in natural language. The data is blended and kept up to date by Coupler.io, a reporting automation platform for business teams. |
40,501–40,550 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.