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 |
|---|---|
| genesys-memory astrix-labs/papez Causal graph memory engine for AI agents. Scores memories using relevance × connectivity × reactivation, connects them in a causal graph, and actively forgets irrelevant ones. 11 MCP tools
including store, recall, search, traverse, and expl | |
| pokeapi-mcp-server astro2024/pokeapi-mcp-server MCP server providing tools to look up Pokémon, moves, abilities, items, locations, evolutions, and more from the PokéAPI via natural language. | |
| MailNet MCP Server astroa7m/mailnet-mcp-server Unified email orchestration server for Gmail and Outlook with agentic tools for sending, reading, searching, and managing emails, enabling assistant-driven email workflows. | |
| mame-mcp astrobleem/mame-mcp Enables MAME reverse engineering tasks including memory access tracing, input injection, ROM auditing, and IO-port discovery for arcade-to-SNES porting. | |
| scratch-mcp astroblocksmod/scratchmcp Enables editing Scratch .sb3 projects programmatically, including sprites, blocks, costumes, sounds, variables, and online project sharing, plus headless VM execution and live-reload in TurboWarp Desktop. | |
| threads astroboyfromspace/threads-mcp Enables AI agents to access one social-media account for publishing, replying, reading feeds, profiles and comments, searching, and performing likes, follows, and reposts, using official APIs where available and a persistent remote browser | |
| MCP DuckDuckGo astrocorp/mcp-duckduckgo Enables AI assistants to search the internet using DuckDuckGo and extract clean, formatted content from web pages. | |
| MCP-SQL-Query astrocorp/mcp-sql-query Enables clients to run raw SQL against MySQL, MariaDB, and PostgreSQL, with read-only connections limited to read queries and other connections allowed to execute writes. | |
| MCP Cloud astrodevit-creator/mcp-cloud Enables connecting multiple business tools behind a single managed MCP endpoint with per-connector permissions, dispatch, and audit records. | |
| mcp-server-skeleton astrodevit-creator/mcp-server-skeleton Enables registering custom agent tools via a minimal JSON-RPC over stdio implementation, with zero external dependencies, to expose them to AI agents like Claude, Cursor, and Gemini. | |
| SkyFi MCP Server astrogirlnim/skyfimcp An MCP server providing AI agents with access to satellite imagery, geospatial intelligence, and billing management through the SkyFi platform. | |
| Security Weekly MCP astroicers/security-weekly-mcp An MCP server that generates security weekly reports and manages a terminology glossary for the Taiwanese cybersecurity community. | |
| PyGithub MCP Server astromined/pygithub-mcp-server Enables AI assistants to interact with GitHub through the PyGithub library, providing tools for managing issues, repositories, pull requests, and other GitHub operations with intelligent parameter handling and error management. | |
| astro-airflow-mcp astronomer/astro-airflow-mcp An MCP server that enables AI assistants to interact with Apache Airflow's REST API for DAG management, task monitoring, and system diagnostics. It provides comprehensive tools for triggering workflows, retrieving logs, and inspecting syste | |
| catia-v5-mcp astroqueststudio/catia-v5-mcp Enables AI agents to drive CATIA V5 through COM for batch part, assembly, and drawing work, validating every step against tool schemas before touching CATIA and verifying results by volume change, 3D geometry references, and final poses. It | |
| thonny-ai astroqueststudio/thonny-ai Enables AI agents to program MicroPython boards such as Raspberry Pi Pico and ESP32 through Thonny, with visible shell commands and optional human approval. | |
| qa-screens astuanax/qa-screens Enables AI-driven visual QA by capturing web pages and comparing screenshots with SSIM to detect visual regressions, changed regions, and previews, while supporting authenticated sessions and OAuth for apps behind login. | |
| agentic-house-search asturksever/agentic-house-search-data-registry Enables AI agents to retrieve a full UK postcode report, perform fast geography lookups, compare up to five postcodes side by side, and search or fetch the underlying open dataset registry through the Model Context Protocol. | |
| MCP Data Science astyanm/mcp-data-science Provides 102 tools for end-to-end data science pipelines, enabling data loading, cleaning, visualization, modeling, and reporting via natural language. | |
| Familiar astynomus-llc/familiar Enables AI agents to operate self-hosted Discord bots with per-session active identity and per-call overrides, covering messages, channels, threads, members, roles, moderation, automod, webhooks, events, and raw REST passthrough. Supports s | |
| QGIS 4 MCP Server astyyym/qgis-4.0-mcp-public Connects AI assistants (Claude, Cursor, etc.) to QGIS 4.x via MCP, enabling natural language-driven GIS workflows through 37+ tools, including data loading, processing, and spatial analysis. | |
| GDP MCP Tools asuar16/gdp-mcp-tools MCP server for Grubhub Data Platform operations, providing 40+ tools across 18 service categories for data platform management, observability, analytics, and collaboration. | |
| Rohlik MCP Server asuka1902/rohlik-mcp Enables AI assistants to search grocery products, manage shopping carts, access account and order information, and find deals and delivery details across several European online grocery platforms. | |
| Korean DART MCP asuka2013/korean-dart-mcp Enables AI assistants to query Korea's OpenDART electronic disclosure system through 15 MCP tools covering filings, financials, ownership, XBRL, and attachments. Adds Buffett-style analyst frames (insider clustering, accounting-risk scoring | |
| Zendesk MCP Server asukaalves1873/zendesk-mcp-server Enables retrieving and managing tickets and comments, analyzing tickets, drafting responses, and accessing Help Center articles as a knowledge base. | |
| FirstData MCP asukafrancois58/firstdata Enables AI agents and users to turn natural-language questions into authoritative primary-source recommendations, authority-level comparisons, and step-by-step website or API navigation paths for retrieving traceable original data. | |
| directus-safe-mcp asukakimya/directus-mcp A schema-aware, dry-run/verify-guarded MCP sidecar that wraps the Directus REST API with safety features like collection allowlists, batch limits, and mutation verification, enabling secure and LLM-friendly data operations. | |
| Codex Control Plane MCP asukascott/codex-control-plane-mcp Enables durable, retry-safe automation of long-running Codex Desktop tasks through MCP, with async operations, polling, Plan Mode workflows, approvals, diagnostics, and restart recovery. | |
| screenshot-website-fast asukascott/mcp-screenshot-website-fast Enables users to capture web pages or specific DOM elements as optimized images, automatically tiling full pages into 1072x1072 chunks, and to record timed screencasts or animated WebP files for AI vision workflows. | |
| Niblet asukauchida34142/niblet-skill-mcp An MCP server that lets coding agents ground UI work in real product screens by fetching one to three reference screens as inline images for a specific layout, state, or interaction question, and by looking up fonts, icons, or animated icon | |
| AWS Cost Explorer and Amazon Bedrock Model Invocation Logs MCP Server asukav03/aws-cost-explorer-mcp-server Enables users to analyze AWS spend and Bedrock model usage through natural-language queries, including EC2 costs, service spend reports, and breakdowns by region, model, and user. | |
| OfficeReader-MCP asunainlove/officereader-mcp Converts Microsoft Office documents (Word, Excel, PowerPoint) to Markdown format with intelligent image extraction and optimization, preserving document structure and metadata. | |
| file-analyzer-mcp asuramama-hue/fileanalyzer_mcp_testmonial Local MCP server that extracts text and structure from PDF, DOCX, PPTX, SVG, and PNG files, along with directory and metadata listing, for read-only document analysis by LLM hosts. | |
| weekly-verify asuramama-hue/weekly-verify-mcp Compares weekly work reports against original data to identify incorrect values, missing items, and narrative inconsistencies, providing verification results to users. | |
| OpenDota MCP Server asusevski/opendota-mcp-server Model Context Protocol server that enables LLMs and AI assistants to retrieve real-time Dota 2 statistics, match data, player information, and game metrics through a standardized interface. | |
| Workspace Bridge asutorufa/workspace-bridge Enables ChatGPT to access and manage registered local development projects through controlled file, command, process, and Git operations, with batch tools, permission modes, and tunnel-based remote forwarding. | |
| mcp-homelab asvarnon/mcp-homelab An MCP server that gives AI assistants real-time access to your homelab infrastructure. It enables querying node status, managing Docker containers, controlling Proxmox VMs, and inspecting OPNsense firewall state through natural conversatio | |
| Verified Technical Solver asvlcii/verified-technical-solver Provides deterministic tools for symbolic math, equation solving, relation verification, reasoning structure audits, and validated CAD primitives, returning evidence artifacts for every result. | |
| splitsmart-mcp aswanisahoo/splitsmart Enables group expense management and fair settlement via MCP, allowing creation of groups, recording expenses, viewing balances, and computing minimal transfers to settle debts. | |
| TradeMCP aswathas/trademcp Enables AI agents to research Indian equities (NSE/BSE) with 14 read-only tools covering live quotes, OHLCV history, 57 technical indicators, fundamental ratios, financial statements, stock screening and portfolio analytics. Everything runs | |
| 2021 Roblox Studio MCP Bridge aswdffdas-sudo/octane-mcp Enables AI coding assistants to connect to 2021 Roblox Studio / Aisaka Studio to capture screenshots, execute Luau, inspect and modify the DataModel, read/write scripts, and retrieve output logs via MCP. | |
| KiCad MCP Server aswego123/kicad-custom-mcp A Model Context Protocol (MCP) server that exposes KiCad PCB design automation tools to AI assistants and other MCP clients. | |
| indian-stock-market-mcp aswin-s/indian-stock-market-mcp MCP server providing real-time Indian stock market data through 15 tools, including stock details, historical data, market overview, and news. Works with Claude Desktop, Claude Code, Cursor, Windsurf, and other MCP clients using an API key | |
| jira-mcp-server aswin-sankaranarayanan/jira-mcp-server Exposes JIRA sprint data as MCP tools and prompts, with mock data and a pluggable client for real JIRA integration. | |
| Banking MCP Server aswin922/banking-mcp-server Enables AI assistants to perform banking operations like creating customers and accounts, checking balances, transferring funds, and viewing transaction history through MCP tools. | |
| Kafka MCP Server aswinayyolath/kafka-mcp-server A comprehensive MCP server for Apache Kafka operations, enabling seamless integration with Claude Desktop and other MCP clients for cluster management, topic operations, and message streaming. | |
| MCP Weather & Accruals Server asxrnd/mcp_deepseek Provides weather data via OpenWeatherMap API, file operations, and integration with remote PostgreSQL database for managing accrual records through a FastAPI backend, all accessible through natural language via Deepseek. | |
| Wheelman asymmetra/rekt-wheelman Exposes Rekt Knockout and Minute trading over stdio as MCP tools: list markets, preview stake and payout after fees, and check open positions, plus opening and closing positions when a scoped session key or paper mode is configured. Signed | |
| binance-mcp-demo asyn18/binance-mcp-demo A local stdio MCP server for Binance Spot, enabling users to query prices, manage orders, and check account info via natural language through the Hermes agent. | |
| Pierre Fitness Platform MCP Server async-io/pierre_mcp_server Connects AI assistants to fitness data from over 150 wearables including Strava, Garmin, and Fitbit through the Model Context Protocol. It provides 47 tools for sports science-based analysis, training load management, recovery tracking, and |
31,401–31,450 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.