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 |
|---|---|
| mcp-image daikileone60/mcp-image Enables users to generate and edit images from MCP-compatible clients, with automatic prompt optimization and quality presets, and supports Gemini, OpenAI GPT Image, and BytePlus Seedream. | |
| swissdevjobs daikipanagiotou/swissdevjobs-cli Enables AI agents to search, filter, and apply to salary-transparent tech jobs across nine international job boards, including Switzerland's jobs.ch and jobup.ch and Singapore's MyCareersFuture. Results are returned as JSON with salary, rem | |
| Pentest MCP daikiward0871/pentest-mcp Enables authorized penetration-testing workflows through MCP by exposing tools for network scanning, web fuzzing, subdomain enumeration, vulnerability scanning, brute-force checks, packet capture, and report generation with HTTP/STDIO/SSE t | |
| Notes MCP Server daikw/mcp-server-on-raspi A simple note storage system that lets users create and summarize notes with varying detail levels using Claude. | |
| Omega API MCP dailui/omega-api-mcp It gives AI agents one tool to access the Omega-API judge, which evaluates conversational turns and returns a decision and gate action (release, clarify, suppress, or none) to guide the agent's response. | |
| Agent Social MCP daily-ac/agent-social-mcp A social protocol for AI agents to communicate with each other through a central Hub, enabling messaging, online status, and broadcasts. | |
| OpenCU daily-ac/opencu Computer use for any AI agent on macOS, Windows and Linux. It reads apps through the accessibility tree and screenshots, acts on elements by index, runs locally with no account, and the person at the keyboard can stop it at any time. | |
| parsely-mcp dailybeast/parsely-mcp MCP server that enables AI assistants to query Parse.ly analytics for top posts, authors, tags, referrers, content search, and social shares. | |
| Logseq MCP Server dailydaniel/logseq-mcp A server that enables LLMs to programmatically interact with Logseq knowledge graphs, allowing creation and management of pages and blocks. | |
| Playwright Browser MCP Server dailynobug/playwrightbrowsermcpserver A human-in-the-loop browser MCP server that pauses for user actions like login or CAPTCHA, saves and reuses authentication contexts, and extracts page data for development assistance. | |
| recount dainemawer/recount Enables users to ask about render performance in React and React Native apps via natural language, exposing tools for listing render hotspots and explaining component re-renders. | |
| MCP-researcher Server dainfernalcoder/perplexity-mcp This TypeScript-based server implements a simple notes system using MCP concepts, enabling users to create, list, and summarize text notes through natural language prompts. | |
| Daipendency daipendency/daipendency-mcp Get the narrative and API documentation for the exact version of any of your dependencies. (Only Rust is supported at the moment.) | |
| worldanvil-mcp-server daireb/worldanvil-mcp MCP server for World Anvil that enables browsing, searching, and managing world content, plus exporting complete copies for migrations or backups. | |
| Papernote MCP Server daishir0/papernotemcp Enables Claude.ai to interact with the Papernote cloud-based note management system to create, read, and manage notes and research papers. It supports operations like text replacement, content appending, and paper summary retrieval through | |
| Lucid Observability Agent daishizensensei/lucid-observability-agent MCP server for monitoring, diagnosing, and auto-correcting observability issues across platform services, integrating with Sentry and PostgreSQL for error tracking and billing pipeline health. | |
| pptx-minio-mcp daishuaishuai/pptx-minio-mcp Enables PowerPoint presentation generation and uploading to MinIO storage, returning presigned download links. | |
| Google Tasks MCP Server daismond/mcp_task_google Integrates with Google Tasks to allow listing, reading, searching, creating, updating, and deleting tasks. | |
| desmume_webassembly_harness daisukedaisuke/desmume_webassembly_harness Enables Codex to control a DeSmuME WebAssembly emulator for Nintendo DS games, including state management, screenshots, and persistent scripting via stdio MCP. | |
| local-mcp-chatgpt-tunnel daisukedaisuke/local-mcp-chatgpt-tunnel Enables ChatGPT Developer Mode to securely invoke multiple Windows-local stdio MCP servers through a single gateway, with tool namespacing, path restrictions, and serial execution control. It includes safe-file, image, and download tools fo | |
| OBS Control MCP daisukedaisuke/obs-control Enables control of OBS Studio via MCP, including media source playback, scene management, audio mixer adjustments, and screenshot capture, using OBS's WebSocket v5 protocol. | |
| Firecrawl MCP Server daisukehori/firecrawl-mcp Enables Claude to scrape and crawl websites via a self-hosted Firecrawl instance through the Model Context Protocol. Provides tools for single/multi-URL scraping, site mapping, and full-site crawling operations. | |
| Gemini Image MCP daisukehori/gemini-image-mcp Enables AI-powered image generation and editing using Gemini and Imagen models, supporting text-to-image, image editing, and multi-image composition through MCP tools. | |
| Yamato Printer MCP Server daisukehori/yamato-printer-mcp-server Enables printing of Yamato B2 Cloud shipping label PDFs to a thermal label printer through a Raspberry Pi-based MCP server. | |
| fieldroutes-mcp daisydaines/fieldroutes-mcp Read-only MCP server enabling pest and lawn owners to ask FieldRoutes questions in plain English, with owner-shaped tools and a hosted connect vault. | |
| fieldwork-mcp daisydaines/fieldwork-mcp Enables pest control business owners using Fieldwork to ask plain-English questions about customers, invoices, product usage, schedules, and technicians, returning clear answers from the Fieldwork API. Read-only access with no ability to mo | |
| Ask GorillaDesk (gorilladesk-mcp) daisydaines/gorilladesk-mcp Read-only MCP server that lets Claude, ChatGPT, or Cursor answer pest-business questions from your GorillaDesk account in plain English—looks up customers, invoices, schedules, and technicians without creating or changing data. | |
| research-project-mcp-demo daisylandforu/research-project-mcp-demo A demo MCP server for research project management that enables document ingestion, information extraction, project tracking, and report generation through MCP tools. | |
| postgres-mcp daitasu/postgres-mcp Enables read-only SQL query execution and database schema information retrieval for PostgreSQL databases. | |
| Odoo MCP Skeleton daitianjun/odoo-mcp-demo A Python-based MCP server for interacting with Odoo ERP, offering example tools such as health checks, customer search, customer details, and listing sale orders. | |
| db-connector daix1999/db-connector Enables AI agents to connect to and interact with MySQL, Redis, and MongoDB through a stdio MCP server, with unified data source discovery, read-only querying, and explicitly authorized write operations. | |
| SpecQ MCP Server daizehua-wq/specq-mcp Enables AI agents to generate structured intelligence reports for electronic chemical sales, including competitor analysis, visit logging, and insight extraction. | |
| Codex OpenViking Memory daizhonggeng/codex-openviking-memory Provides eight scoped, read-only tools for health checks and for finding, searching, reading, listing, traversing, grepping, and globbing a configured user's native memories and personal-context subtree, without exposing capture or write op | |
| android-mcp-harness dajarony/android-mcp-harness Local MCP server for Android emulator control, exposing tools to get emulator status, inspect UI tree, capture screenshots, and open App settings via Appium and ADB. | |
| cadesimu-mcp dajarony/cadesimu-mcp Enables parsing, lossless round-tripping, and generation of CADe_SIMU .cad schematic files, including three-phase power circuits and direct starter control circuits with STOP/START and KM1 self-hold, plus validation status reporting. | |
| mcp-universal-gentes dajarony/mcp-universal-gentes Enables agents to connect and disconnect OpenAPI services, then list, describe, and invoke newly discovered capabilities through a stable six-tool MCP interface without changing the published tool catalog. | |
| midnight-nextjs-mcp dajo-code/midnight-nextjs-mcp Combines Midnight Network blockchain development tools with Next.js DevTools for building decentralized applications, supporting contract development, wallet management, and Next.js diagnostics. | |
| schoology-mcp dajun666/schoology-mcp An MCP server that automates login to PAUSD Schoology via ClassLink and provides tools to fetch grades, upcoming assignments, and recent posts. | |
| decide-mcp dakdevs/decide-mcp Enables agents to delegate decisions to a configurable model, returning recommended choices with percentage scores and profile-based routing. | |
| polygloss dakdevs/polygloss Enables coding agents to read locally submitted code-review comments and diff context over MCP, so they can fix the code and request another review. | |
| brain dakirchik/agent-recall Enables AI agents to search a single local SQLite full-text index of past Claude Code and Codex sessions, Obsidian notes and attachments, GitHub activity, and imported Claude/ChatGPT chats, plus project timelines and recent items. It also l | |
| mcp-after-effects dakkshin/after-effects-mcp A Model Context Protocol (MCP) server for Adobe After Effects that enables AI assistants and other applications to control After Effects through a standardized protocol. | |
| arxiv-mcp dakl/arxiv-mcp Provides tools to search for arXiv papers, retrieve full-text HTML content via ar5iv, and download papers as PDFs. This server enables LLMs to find and interact with scientific research literature through a standardized interface. | |
| custom-mcp-server dakofler/ollama-mcp-demo Exposes custom Python functions as tools and integrates with Ollama for tool calling. | |
| agent-context-router dakotafabro/agent-context-router Enables agents to load instruction modules on demand from linked git repos, keeping the always-loaded AGENTS.md slim. It syncs each module's repo before reading and returns the module text with the commit it was read at. | |
| bpm-mcp dakotafabro/bpm-mcp Enables agents to search and inspect behavioral packages from the BPM registry and install or remove them with mandatory human approval, all without leaving the agent session. | |
| chatgpt-mcp dakotafabro/chatgpt-mcp Enables reading public ChatGPT share links and converting them into markdown transcripts or structured JSON for use in Goose. | |
| mcp-consensus dakraid/mcp-consensus A Model Context Protocol (MCP) server that implements an advisor-based consensus mechanism for collaborative problem-solving using multiple AI models, enabling structured discussion and debate among 5 specialized advisors to reach optimal s | |
| Gemini MCP Server dakrin/mcp-gemini-server A Model Context Protocol server that enables Claude Desktop to interact with Google's Gemini 2.5 Pro Experimental AI model, with features like Google Search integration and token usage reporting. | |
| prforge daksh1403/prforge PRForge is an agentic PR bot that reads a GitHub issue, understands the codebase, writes a fix, runs tests in a sandbox, and opens a human-reviewed pull request. It can also run as an MCP server to provide tools for Claude Desktop or Cursor |
41,701–41,750 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.