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 |
|---|---|
| todo-mcp denerportela/todo-mcp Enables Claude to read, create, and complete tasks in a personal Microsoft To Do account through the Microsoft Graph API, including lists, daily and overdue tasks, due dates, reminders, notes, and checklist steps. | |
| db-mcp denezhkinia/shop-db-mcp Enables AI agents to safely interact with a SQLite shop database through schema discovery, read-only SQL queries, and pre-built analytics reports like top customers, top products, and revenue summaries. | |
| chrome-bridge-mcp deng1986/chrome-bridge-mcp Bridges AI agents (Codex, Claude Code) with a real Chrome session via CDP, enabling Google Search and Google AI interactions through the browser. | |
| everything-mcp dengbojing/everything-mcp Enables fast Windows file and directory lookups through the local Everything index, supporting exact, keyword, wildcard, regex, fuzzy, pinyin, and typo-tolerant searches with filters, pagination, directory browsing, counting, and status che | |
| web-gpt dengfeng-0908/codex-web-gpt Enables Codex to send text tasks to ChatGPT web, await responses, continue conversations, choose reasoning effort, and manage projects through MCP tools on macOS. | |
| email-mcp dengjianbo3/email-mcp A self-hosted MCP server for Gmail and Outlook that lets MCP clients read, search, send, reply, and manage email via OAuth, with no built-in secrets and support for both stdio and HTTP transports. | |
| logic-mcp dengtaowei/logic-mcp Enables MCP clients like Cursor to control logic analyzers, open capture files, decode buses such as Intel 8080, and reconstruct display framebuffers like ST7789. | |
| nano-banana-mcp denialgelon/nano-banana-mcp A minimal MCP server that wraps Google Gemini image generation, enabling users to generate images via a single tool prompt in Claude Code or other MCP hosts, with automatic filename collision handling. | |
| omie-finance-mcp denilsonpy/omie-finance-mcp Exposes OMIE ERP's financial API as MCP tools, enabling AI agents to query bills, process payments, generate boletos/PIX, and read cash flow via natural language. | |
| Tbrain denis-tang/tbrain Enables coding agents to bootstrap, search, save, and resume local Markdown-backed session knowledge across Claude Code, Codex, DSH, and pi, preserving project prompts and task progress across sessions and hosts. | |
| Study Projects MCP denis-zag/study_projects_mcp Provides local MCP tools for searching, filtering, adding, updating, and analyzing a catalog of study projects, with statistics and GitHub publication readiness checks, all backed by SQLite. | |
| social-watch-mcp denis1stomin/social-watch-mcp A read-only MCP server that lets Claude query your own Threads, itch.io, YouTube, and X accounts through official APIs, returning recent posts, per-post stats, replies and mentions, and account totals. It also supports saving daily snapshot | |
| LinkedIn Prospecting CSV Manager denis911/antigravity-mcp-csv-add-deduplicate Manages LinkedIn prospecting CSV files with deduplication and search, reducing LLM token usage by offloading file operations. | |
| Twenty CRM MCP Server denisalmeida/twenty-crm-mcp Enables interaction with Twenty CRM through a Model Context Protocol server. Provides comprehensive CRM operations including managing people, companies, opportunities, notes, tasks, and custom objects with support for filtering, pagination, | |
| searxncrawl denise0022/searxncrawl Enables AI agents to run web searches across SearXNG and/or You.com in parallel and to crawl single pages or whole sites into clean, deduplicated Markdown or JSON, including behind logins. | |
| uisight-mcp denise1917/uisight Enables AI agents to run live mobile and desktop browser sessions side by side and measure real UI properties — contrast ratios, touch-target sizes, overflow, clipped content and light/dark theme drift — returning precise findings as text w | |
| fcp-mcp-server denisebonnet/fcp-mcp-server Enables Claude to read, edit, and generate Final Cut Pro timelines, with cut detection, markers, roles, transcript-based editing, rough cuts, keyframes, and live pushes into a running Final Cut Pro. Every write is journaled with undo suppor | |
| MCP Proxy Server denisekose264/mcp-proxy-server Aggregates multiple backend MCP servers (Stdio, SSE, and Streamable HTTP) and exposes their combined tools and resources through a single unified SSE or Streamable HTTP endpoint, or as a standalone Stdio server. It routes requests to the co | |
| Octopus Deploy Official MCP Server denisekose264/mcp-server Enables AI assistants to inspect, query, and diagnose an Octopus instance and, when write access is enabled, create releases, deploy releases, run runbooks, and adjust feature toggles. | |
| 12306 Train Ticket Search MCP Server deniselewis200081/12306-mcp Enables searching for train tickets on China's 12306 railway system, including ticket availability queries, train filtering, station stopover information, and transfer route planning. | |
| LeetCode MCP Server deniselombardi/leetcode-mcp-server Enables automated access to LeetCode problems, solutions, contests, and public or authenticated user data across leetcode.com and leetcode.cn, including problem search, profile and submission retrieval, code execution, and note management. | |
| whoop-ai-mcp deniselombardi/whoop-mcp Enables users to ask natural-language questions about their WHOOP recovery, sleep, HRV, strain, workouts, and body measurements, with tools for baselines, trends, sleep debt, period comparisons, and weekly health reviews. | |
| : denisemonteiro64811/swissdevjobs-cli Enables: | |
| Perforce P4 MCP Server denisemyers76/p4mcp-server Enables MCP-compatible AI assistants to safely read and write version control data such as files, changelists, shelves, workspaces, jobs, reviews, streams, and server metadata, with optional DAM asset management via REST APIs. | |
| jira-full-mcp denisglazkov/jira-full-mcp A local MCP server for administering Atlassian Cloud Assets and Jira, providing hundreds of tools for schemas, projects, workflows, and more. | |
| mcp-graph-loop denishmistry07/mcp-graph-loop Enables AI agents to orchestrate tasks as a DAG with automated validation loops, executing validation commands, tracking statuses, and allowing iterative code fixes until tasks pass. | |
| MCP GraphQL Server Multi-Fuente denisijcu/deepseek-graphql---mcp-project Enables querying and mutating data across multiple sources (CSV, Google Sheets, SQLite, PostgreSQL, MySQL, MongoDB, Oracle) through a dynamic GraphQL API, with flexible filtering, batching, persisted queries, and dynamic source switching. | |
| Super Calculadora MCP denisijcu/mcp-calculadora An MCP server providing over 30 mathematical and statistical tools, enabling language models to perform calculations, trigonometry, statistics, combinatorics, and list operations through natural language. | |
| tello-mcp denisijcu/mcp-tello MCP server that connects a local LLM to a DJI Tello drone via UDP, enabling natural language flight planning and execution with safety-focused separation of plan and execution, plus a full simulator for development. | |
| plc-mcp denisijcu/plc-mcp An MCP server that connects a local LLM to an industrial PLC via Modbus TCP, enabling process reading, alarm correlation, and safe diagnosis with restricted write access through an allowlist. Includes a simulated PLC with real physics for l | |
| Vertex Cyber MCP denisijcu/vertex-cyber-mcp Enables local AI models to perform defensive cybersecurity analysis through narrowly scoped read-only tools for host posture, Windows security operations, file/IOC triage, code scanning, and allowlisted filesystem/network access while enfor | |
| visioncore denisijcu/vision-core-mcp Enables an AI agent to act as a security operations operator by providing tools to monitor camera feeds, analyze scenes, review situations and alarms, acknowledge or resolve alarms, make voice announcements, and consult audit trails and sys | |
| backloggd-mcp denisionsoft/backloggd-mcp An MCP server for Backloggd, the video game tracker. It lets an AI assistant search the game catalogue and read and manage your library — statuses, ratings, logs, play sessions, reviews and lists — as the signed-in you. | |
| Pactus deniskarlinsky/iso20022-mcp Pactus is an MCP server for parsing and validating ISO 20022 payment messages directly from chat. It exposes nine tools that let AI assistants inspect or validate pacs.008, pacs.002, pain.001, and camt.053 messages — the message types at th | |
| ClickUp MCP Server deniskrds/clickup-mcp-server Enables natural language management of ClickUp workspaces, including task CRUD operations, task listing, and user profile retrieval via Claude Desktop. | |
| loom-fs denislamard/loom-fs-mcp Grants an agent a bounded view of the filesystem where each directory is declared explicitly with a role (reference, sourcecode, workspace) that caps operations to read, write, or delete, resolving every path through a single confinement an | |
| loom-memory denislamard/loom-notes Enables Claude Desktop and compatible agents to search, read, add, update, and delete a locally stored, user-curated memory using hybrid retrieval with reranking and Qdrant. All processing runs locally without cloud services or API keys. | |
| loom-plan denislamard/loom-plan Serves Claude Desktop a unified, actionable read of Google Calendar and Google Tasks, surfacing overdue items first, then the day's events, free slots, and open tasks, with explicit-request-only write tools for creating, updating, completin | |
| Figwright Kiwi Reader denispapushajava/figwright-kiwi-reader Enables Codex and other MCP clients to read a selected Figma frame from an open Chrome tab, exposing decoded design structure and layout data through read-only MCP tools. | |
| india-business-mcp-server denispaul2406/india-business-mcp-server Enables LLMs to query real-time Indian bank branch details, postal PIN codes, validate GSTIN/PAN structures, check e-commerce serviceability, and compute GST breakdowns using free public APIs and offline verification logic. | |
| Pingfusi deniz443/pingfusi It enables AI agents to call human reviewers for second opinions and human judgment on artifacts such as designs, wording, and videos, then iterate using pinned feedback until the work is approved. | |
| Sync Vault MCP deniz9658/obsidian-sync-vault-ce Enables LLMs and AI agents to directly access and operate on files in supported cloud storage, with tools for listing, reading, creating, deleting, moving, renaming, downloading, uploading, semantic search, share links, and URL upload tasks | |
| lattice-mcp denizaoki25809/matryoshka Enables token-efficient analysis of documents far larger than an LLM context window by executing constrained Nucleus commands server-side and returning compact handles instead of full results. Supports document loading, grep/filter/aggregat | |
| memgate denizayhan04/memgate Access control, conflict resolution, and audit for shared agent memory. Policy-gated memory tools over Postgres + pgvector, exposed via MCP. | |
| evds-mcp-server denizcakiroglu322/evds-mcp-server Enables querying Turkish economic data (inflation, FX rates, policy rates, etc.) from TCMB EVDS via curated tools, preventing LLM hallucination. | |
| LINE OA MCP Ultimate denizd93830/line-oa-mcp-ultimate Enables AI agents to operate one or many LINE Official Accounts through natural language — sending broadcasts, Flex/Rich/card messages, building rich menus, coupons, audiences, insights, and LINE Shopping operations, with pre-flight quota c | |
| motherduck-mcp denizguzeloglu/motherduck-mcp Enables executing SQL queries on DuckDB databases locally or on MotherDuck cloud, with support for multiple databases, read-only mode, and Claude Desktop integration. | |
| Charlotte denizkaramanlis35063/charlotte Enables AI agents to browse, navigate, and interact with web pages through structured, token-efficient page representations — returning compact orientations and targeted element queries instead of full accessibility-tree dumps. | |
| Smart Coding MCP denizmartins/smart-coding-mcp Enables AI coding assistants to search codebases by meaning using local AI embeddings, so they can find relevant code even when terminology differs, while also offering package version lookup, indexing, cache, workspace, and status tools. | |
| screenshot-website-fast denizmatsumoto/mcp-screenshot-website-fast Enables users to capture full webpages or specific DOM elements as PNGs, with automatic tiling into 1072x1072 chunks optimized for AI vision models, and to record screencasts or animated WebP files. |
43,601–43,650 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.