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 |
|---|---|
| plane-ce-mcp bl4nk44/plane-ce-mcp Unofficial MCP server for self-hosted Plane Community Edition, enabling project management tasks like managing work items, cycles, modules, and initiatives via natural language. | |
| codex-codebuddy-mcp black-94/codex-codebuddy-mcp Enables Codex to control CodeBuddy Code through the Agent Client Protocol, including launching local or SSH sessions, sending prompts, handling permissions, and managing session lifecycle. | |
| Vision MCP Server black-94/lm-visual-mcp A Model Context Protocol server that provides text-only LLMs with visual capabilities through Z.AI-compatible vision tools, routing requests across multiple providers with automatic fallback. | |
| Gurume black-birthday/gurume-unified Enables AI assistants to concurrently search, normalize, and compare Japanese restaurant results across Tabelog, Hot Pepper Gourmet, and Google Maps, with filters for cuisine, area, reservations, and amenities. It also retrieves structured | |
| ORA black-dash77/ora-mcp Persistent memory + intelligent context for Claude Code. Save and reuse solutions, decisions, and patterns across projects. | |
| qca-agent-mcp black-jony/qca-agent-mcp Exposes eight QCA (quantum calibration) tools via a standards-based MCP stdio server for MCP-compatible agents to interact with backend status, experiments, schemas, parameters, dry-runs, history, and arrays. | |
| Car Deals Search MCP black-table/auto_search_mcp Enables AI assistants to search and aggregate used-vehicle listings across Cars.com, Autotrader, and KBB, with filters for price, mileage, dealer information, and CARFAX-style history conditions. | |
| @ethioviral/mcp black12-ag/ethio-viral-mcp Enables browsing Ethio-Viral's live catalog and placing orders for gift cards, subscriptions, and SMM services from any MCP-capable client using a single API key. | |
| gpt-cross-validator black940514/gpt-cross-validator Enables cross-validation of Claude's plans and code reviews using GPT-5.2, providing independent assessment with structured JSON output. | |
| ssh-mcp blackaxgit/ssh-mcp Enables AI assistants to execute shell commands and transfer files via SFTP across remote servers using existing SSH configurations. It supports parallel execution on server groups and provides built-in safety warnings for potentially destr | |
| AdsCraft MCP blackbeans-1301/cocos-mcp An MCP server for Cocos Creator 3.8.8 that lets AI assistants read and drive the project open in the editor, covering scene inspection, node and component creation and editing, prefab and asset management, script execution and builds. It ex | |
| Blackbird MCP Server blackbirdai-team/blackbird-mcp-server Enables LLMs to access Blackbird AI's services through the Model Context Protocol, using Blackbird API credentials for authentication. | |
| PostgreSQL Remote Cloudflare MCP Worker blackboxprojects/postgres-remote-mcp A Cloudflare Worker that provides a Model Context Protocol (MCP) interface for PostgreSQL databases, allowing AI assistants like Claude and Cursor to interact with PostgreSQL databases through natural language. | |
| Multi-Browser MCP blackcatofficialytb/multi-browser-mcp MCP server for multi-engine browser automation with proxy rotation, enabling AI-driven web interaction and task execution across Camoufox, DrissionPage, and Browser Use. | |
| Xbitto MCP blackcheet/xbitto-mcp Lets AI agents work inside a creator account to create short links and paywall links, fill access code pools in bulk, and upload galleries, videos, and multi-episode series straight from the local computer to the user's own Bunny storage. U | |
| Meiju MCP blackcocwhite/meiju-mcp Enables an AI agent to sign in to a user's own Midea air-conditioner cloud account through a local browser page, discover bound devices, read status and capabilities, and send experimental power, temperature, mode and fan-speed controls wit | |
| ebpf-mcp-tracer blackdragoon26/ebpf-mcp-tracer Enables LLMs to safely write and run bpftrace scripts against the Linux kernel for observability, with explicit probe allowlists and execution timeout. | |
| Mobix blackfoxxx/mobix Authorized Android pentest lab, drivable by hand or by AI — Frida bypass chain, live traffic capture with automatic IDOR flagging, full MCP control plane, web dashboard. | |
| Discord Community MCP blackgirlbytes/discord-community-mcp Enables AI agents to search and recall Discord conversations, analyze community activity, participation, and growth, and optionally send messages when writes are enabled. | |
| GitHub Calendar MCP Server blackgirlbytes/github-calendar-mcp-server Enables AI assistants to interact with GitHub project boards as calendar events, analyze team workloads, and provide scheduling insights. Supports team status tracking, workload distribution analysis, and smart task assignment recommendatio | |
| mcp2mqtt blackhker/mcp2mqtt This server bridges the Model Context Protocol (MCP) to the MQTT protocol, allowing LLMs to interact with IoT devices and hardware systems through configurable message topics. It enables users to map natural language prompts to specific MQT | |
| HubSpot MCP Server blackhole2100/ai-assistant-with-hubspot-crm Enables AI assistants to interact with HubSpot CRM data, providing tools for managing contacts and companies, retrieving activity and conversations, and performing semantic search with vector storage and caching for improved performance. | |
| urlscan-mcp blackhole8080-darkmatter/urlscan-mcp MCP server for urlscan.io that scans URLs, searches historical scan data, and assesses indicators with compact, context-efficient summaries instead of raw API responses. | |
| Blackhole MCP Server blackholedex/mcp Model Context Protocol server for the Blackhole DEX on Avalanche, providing 32 tools for swaps, liquidity, CL positions, voting, locks, gauges, bribes, yield queries, vote leaderboard, epoch state, allowances, pool status, risk flags, and t | |
| Africa’s Talking MCP blackie360/africastalking-mcp Enables AI clients to check an application balance and preview or send bulk SMS and airtime, either through a hosted OAuth connection that needs no local install or via a self-hosted stdio/Cloudflare Workers deployment using your own creden | |
| Luma Events MCP Server blackie360/luma-events-mcp An MCP server for managing Luma events, including listing, creating, updating events and guest registrations with privacy-conscious summaries. | |
| aigoodbaiduyun blackielf/aigoodbaiduyun Enables AI assistants to manage Baidu Netdisk through local login: QR-code sign-in, listing, creating, moving, renaming and copying files, plus background uploads and downloads. Transfer tasks persist across sessions with pause/resume and r | |
| Custom ClickUp MCP blackklegend/custom-clickup-mcp Local MCP server for essential ClickUp workflows over stdio, exposing a bounded set of tools for search, task management, comments, tags, and workspace hierarchy. | |
| Postgres MCP Server blackklegend/postgres-mcp Enables interaction with PostgreSQL databases through natural language commands. Supports schema exploration, table inspection, DDL generation, data preview, and safe SQL execution with built-in query limits. | |
| agent-browser-mcp blackman99/agent-browser-mcp Exposes the local agent-browser CLI as MCP tools for browser automation and provides one-command integration with Codex. Enables navigation, interaction, page reading, tab management, and session control through natural language commands. | |
| airing blackman99/airing Checks which .env, .git, keys and source maps Cloudflare Pages, Vercel, Netlify or Firebase would publish, before and after deploy. CLI, live probe, Claude Code hook, MCP server and GitHub Action. | |
| cronlag blackman99/cronlag Shows, cron slot by cron slot, how late each scheduled GitHub Actions run started, which slots never ran, and whether your config or GitHub's scheduler is to blame; can catch up missed slots without double-running. | |
| failshut blackman99/failshut Tell you, in one command, whether your AI coding-agent permission hooks and deny/ask rules actually fail shut — or silently fail open. | |
| keyroll blackman99/keyroll Find every copy of your AI provider API keys on your machine, see which still work, and rotate a key everywhere at once. Read-only MCP: key_map, key_copies. | |
| rackhands blackman99/rackhands Safe, vendor-neutral server BMC access over Redfish (Dell iDRAC, HPE iLO, Lenovo XCC, Supermicro) with IPMI fallback. Reads by default; power, boot, firmware and virtual media need two opt-in keys, confirmation and an audit record. | |
| spillscan blackman99/spillscan Tell you which secrets spilled into local AI agent transcripts — and scrub them in place with a backup. | |
| spoilheap blackman99/spoilheap Find the disk clutter and leftover processes coding agents leave behind (worktrees, crash dumps, old sessions) and safely clean only what's SAFE. | |
| undomap blackman99/undomap Which paths can your AI coding agent erase with no way back? One read-only command maps them as flood zones, ranked by bytes, with the cheapest fix. CLI + GitHub Action + MCP. | |
| wheelripe blackman99/wheelripe Before you move to Python 3.15 (or 3.15t), find every dependency with no wheel for it on your platforms — and the bump that fixes it. CLI, GitHub Action, MCP server. | |
| Function Concept MCP blackmarketing2026/function-concept-mcp Enables access to a local Markdown knowledge base via MCP, currently only providing a ping tool. | |
| Portal MCP Server blackmatter-studios/portal-mcp Enables users to connect their music distribution catalog and streaming analytics to Claude Desktop, Cursor IDE, and ChatGPT for querying stream trends, inspecting catalog compliance, and applying safe draft release fixes via natural langua | |
| blackmount-mcp blackmount-ai/blackmount-mcp Enables AI assistants to access and search local browser history, bookmarks, open tabs, and downloads for personalized context. | |
| blackmount-nlp-mcp blackmount-ai/blackmount-nlp-mcp NLP without the bloat — 45 tools for sentiment, keywords, readability, summarization. Zero heavy dependencies. 42KB wheel. | |
| seshat blackopsrepl/seshat Records a screen or region into a silent video, ingests timeline events from other tool servers, then synthesizes, aligns, and muxes narration with captions over the take. | |
| LinuxServer MCP blackoutsecure/linuxserver-mcp Enables users and AI assistants to search LinuxServer.io images, audit Compose and docker run configurations, research image requests, validate service specifications, generate Dockerfile/Compose/README scaffolds, and access guidance and be | |
| LinuxServer.io MCP Registry blackoutsecure/linuxserver-mcp-registry Enables searching LinuxServer.io images, auditing Docker Compose and run commands against documentation-derived conventions, and searching open/closed image requests to prepare drafts without submission. | |
| laya-mlx-mcp blackphi6/laya-mlx-mcp Enables typed decisions (choice/score/noul) from Laya-MLX over MCP on Apple Silicon, without text generation. | |
| lume-vnc-mcp blackphi6/lume-vnc-mcp Enables an AI agent to see and control the screen of an already-running macOS virtual machine over VNC, exposing four tools: screenshot, click, type_text, and key_press. This lets the agent drive a disposable VM's UI directly with its own m | |
| migas-mcp blackpilledsoftware-prog/migas-mcp Enables AI assistants to query local Migas meeting transcripts, including searching, listing meetings, and retrieving speaker contributions, all read-only. | |
| UnrealMCPHub blackplume233/unrealmcphub Central management platform for Unreal Engine MCP instances that bridges AI agents with Unreal Engine, handling plugin installation, project compilation, editor launch, crash recovery, and transparent proxy of tool calls. |
34,551–34,600 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.