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 |
|---|---|
| ssh-bridge-mcp berixoo/ssh-bridge-mcp An MCP server that enables LLMs to operate on remote Linux machines via SSH, providing tools for running commands, file transfer, and background process management while keeping credentials hidden. | |
| threejs-devtools-mcp berk08018/threejs-devtools-mcp Enables AI agents to inspect and edit Three.js scenes, materials, shaders, lights, animations, and performance in real time, preview changes live in the browser, capture frames, and turn approved changes into code. It supports vanilla Three | |
| Bitbucket MCP berk19060/bitbucket-mcp Enables AI assistants to interact with Bitbucket Cloud and Server APIs, including listing and retrieving repositories and fetching, creating, and viewing pull requests. | |
| Claude Telegram Supercharged berk19060/claude-telegram-supercharged Enables users to drive Claude Code around the clock from Telegram with voice notes in both directions, memory that survives restarts, scheduled reminders, calendar access, and live visibility into every tool call as it runs. It also keeps s | |
| mcp-discord-mod berkal749/mcp-discord-mod Enables Discord server moderation actions including kick, ban, timeout, role management, channel operations, audit logs, and member management through MCP tools. | |
| actions-guard-mcp berkantacun/actions-guard-mcp Scans GitHub Actions workflow files for dangerous triggers, template injection, unpinned actions, excessive permissions, and secrets in shell commands before they are committed, exposing the checks as MCP tools for agents. | |
| bandlab-mcp berkantacun/bandlab-mcp Enables AI models to read and edit your own BandLab songs via natural-language prompts, performing concrete mix operations like adjusting track levels, pan, effects, and arrangement while creating new non-destructive revisions. | |
| fsguard-mcp berkantacun/fsguard-mcp Enables safe filesystem and git operations confined within a single allowed directory tree, using symlink-resolved path containment to protect against path traversal. Users can read, write, list, and search files, as well as initialize repo | |
| pg-guard-mcp berkantacun/pg-guard-mcp Enables read-only PostgreSQL interaction through safe query execution, query planning, table and column inspection, and privilege checks, with multi-layer defenses against write operations and transaction-control exploits. | |
| redis-guard-mcp berkantacun/redis-guard-mcp Enables safe, read-only interaction with Redis databases via cursor-based tools for retrieving keys, hashes, lists, sets, and sorted sets, backed by ACL DRYRUN-verified privilege checks. | |
| lucius berkantay/lucius Enables AI agents to render and interact with rich HTML one-pagers on a local canvas, with tools for rendering, commenting, selection, and status queries. | |
| PubMed berkaycelk/pubmed-mcp Enables biomedical literature research with 17 tools for searching and retrieving publications, full texts, MeSH terms, citations, clinical trials, drug safety data, and more from multiple sources without requiring API keys. | |
| WikiPeatia MCP Server berkaycelk/wikipeatia-mcp Enables AI assistants to search and read the full Ray Peat corpus and WikiPeatia wiki, including articles, books, interviews, newsletters, and forum posts, with full-text, semantic, and LLM-powered Q&A tools, plus access to Ray Peat's origi | |
| mcp-content-pipeline berkayildi/mcp-content-pipeline Analyzes YouTube videos and X (Twitter) feeds to extract transcripts, generate takeaways, TLDRs, social hooks, and comic-style infographics, and syncs results to GitHub, all callable as MCP tools. | |
| mcp-llm-eval berkayildi/mcp-llm-eval A local MCP server that packages LLM evaluation gates as reusable CI/CD primitives, enabling AI agents to run datasets against models, score responses, and enforce quality thresholds. | |
| mcp-tfstate-reader berkayildi/mcp-tfstate-reader A local MCP server that parses Terraform .tfstate files and lets AI agents audit enterprise infrastructure for security misconfigurations without requiring direct cloud credentials. | |
| turhal-python-runner-mcp berkayturhal38-cloud/turhal-python-runner-mcp Executes Python code in isolated subprocesses with resource limits and timeout, designed for safe execution in Docker containers. | |
| Kontrol Freek berkbayri/kontrol-freek-mcp An MCP server that acts as an AI assumption firewall, intercepting risky decisions, detecting contradictions, and routing human approval requests via Telegram or Slack. | |
| falai-mcp-server berkbirkan/falai-mcp A FastMCP server that exposes core fal.ai model API operations, enabling model catalogue browsing, search, schema retrieval, inference, queue management, and CDN uploads through natural language. | |
| repo-seatbelt berkcangumusisik/repo-seatbelt Runtime safety guardrails for AI coding agents. Checks file access, validates shell commands, and scores your repo's AI safety — all via MCP. | |
| mcp-image berkcaruso/mcp-image Enables AI assistants to generate and edit images from natural-language requests, automatically enhancing prompts and applying quality/provider presets across Gemini, OpenAI, and BytePlus Seedream. | |
| Hukuk MCP berkebeydag/hukuk-mcp Connects AI clients to public Turkish legislation and court/board decisions, letting users search and retrieve legal texts, case law, and institutional rulings, and generate petition drafts and UYAP UDF files. | |
| MiniMax MCP JS berkkonstantinou447/minimax-mcp-js Enables MCP-compatible clients to generate images and videos, synthesize speech, clone voices, and create custom voices through AI APIs. | |
| Teams MCP berkowens/teams-mcp Enables AI assistants to interact with Microsoft Teams and Microsoft Graph APIs for messaging, search, user and team management, and file handling. | |
| YouTube Transcript Remote MCP Server berkshirecorporate/youtube-transcript-mcp Enables Claude AI and other MCP clients to retrieve transcripts from YouTube videos in multiple languages and URL formats, with zero local setup through a hosted cloud server or a deployable Cloudflare Worker. | |
| mcp-virustotal berkwalsh/mcp-virustotal Gives AI agents access to the VirusTotal v3 API for analyzing files, URLs, IPs, and domains, returning detection reports and analyst-pivot relationships such as contacted hosts, dropped files, resolutions, and threat actors. It also support | |
| Binary Reader MCP berlinbra/binary-reader-mcp A Model Context Protocol server for reading and analyzing binary files, with initial support for Unreal Engine asset files (.uasset). | |
| BlueSky MCP Server berlinbra/bluesky-mcp Facilitates interaction with the BlueSky social network via its official API, allowing users to retrieve detailed user profiles and following lists with authentication and error handling capabilities. | |
| News API MCP Server berlinbra/news-api-mcp A Model Context Protocol (MCP) server that provides access to global news articles through the News API. It implements a standardized interface for searching news articles, retrieving top headlines, and listing available news sources. | |
| PolyMarket MCP Server berlinbra/polymarket-mcp Enables interaction with the PolyMarket API to fetch prediction market data such as market information, prices, and historical data, with robust error handling and rate limit management. | |
| EVE ESI Tool berman510/eoe_mcp An MCP server that integrates EVE Online's ESI API with AI assistants to enable natural language interaction with character data, market information, and assets. It provides 22 tools for managing skills, wallet balances, ship fittings, and | |
| ChatGPT MCP Server bermingham85/chatgpt-mcp-server Enables ChatGPT to perform filesystem operations, execute PowerShell commands, and manage n8n workflows on a Windows machine. | |
| MCP Puppet Pipeline bermingham85/mcp-puppet-pipeline Enables browser automation and AI-driven workflows through Puppeteer integration, with file management, agent coordination, and QNAP network storage support. | |
| uisight bernard1381/uisight Enables a developer and their AI agent to share the same live mobile and desktop browser sessions, letting the agent navigate pages and receive measured UI findings — contrast ratios, undersized touch targets, overflow, and theme drift — as | |
| duckdb-codegraph bernard357/duckdb-code Exposes three read-only tools—impact, explain, and report—so an agent can query a prebuilt local DuckDB graph of a codebase to find the transitive reverse-dependencies of a symbol or file, inspect a symbol's structural neighbourhood, and in | |
| iLEAP MCP bernard357/ileap-mcp Enables AI assistants to query product footprints, inspect transport activity data, and submit events from iLEAP-conformant endpoints through natural language, using MCP tools over Streamable-HTTP or stdio. | |
| Cinema4D MCP Server bernard4972/cinema4d-mcp Enables Claude to drive a 3D modeling application through natural-language prompts for scene creation, object modification, materials, lighting, animation, MoGraph, dynamics, and rendering. | |
| SerpApi MCP Server bernard89768/serpapi-mcp Enables AI assistants to run search queries across Google, Bing, Yahoo, DuckDuckGo, YouTube, eBay and many other engines, returning structured JSON or compact Markdown for organic results, news, images, shopping, weather, and stock data. It | |
| DataClaw MCP Server bernardcaldas/dataclaw-mcp AI-first CSV analysis tool that enables AI agents to analyze, query, and audit large CSV files directly within conversations, turning raw data into actionable insights. | |
| OpenClaw MCP Server bernarddaniels/openclaw-mcp Enables Claude clients to securely connect to self-hosted OpenClaw assistants via OAuth 2.1, exchange chat messages, check gateway health, manage asynchronous tasks, and route requests across multiple instances. | |
| Project OS for Codex bernardfernandez28/project-os-for-codex Enables Codex sessions to read a project's kickoff card, acceptance criteria, handoff package, progress, Git trail, and next step, then append validated, idempotent progress events backed by matching project-record Git commits. Access is li | |
| Pingfusi bernardgrigorev94318/pingfusi An MCP server that lets AI agents file review jobs with real human reviewers, who open the published work and return pinned comments anchored to the exact elements that are off. Those comments land back in the agent's context so it can revi | |
| oh-my-cassette bernardjohansson57/oh-my-cassette Turns natural-language requests into finished, beat-synced montage videos by analyzing raw clips, selecting shots, matching and syncing music, adding subtitles, transitions, and picture-in-picture, then rendering a final cut for review and | |
| LiDAR Harness MCP bernardleex526-png/lidar_harness_mcp An incremental validation engine that acts as a pluggable MCP middleware for code agents like Claude Code and OpenCode, reducing context overhead by 60-80% through a PGO-inspired four-layer verification architecture. | |
| comfyui-mcp bernardogv/comfyui-mcp MCP server that lets AI agents drive a remote ComfyUI instance end-to-end, including queueing workflows, viewing outputs, downloading models, and authoring graphs via ComfyScript. | |
| Gate22 bernardr915/gate22 Enables teams to govern which MCP tools agents can use, what they can do, and how it’s audited by onboarding remote MCP servers, setting credential modes and function-level allow lists, and exposing permitted MCP configurations through a un | |
| Image Context Runtime bernardroux67868/codex-image-context-runtime Enables Codex to run image generation and inspection as durable local jobs while keeping provider-returned media outside MCP results and returning only bounded text, hashes, relative references, and Job IDs. | |
| Agent Search MCP bernardz49176/agent-search-mcp Enables English and Chinese web search starting without any API key, aggregating zero-key and optional paid providers into compact, multi-source evidence packets. Exposes provenance, provider families, partial failures, stop reasons, and re | |
| taskops bernatch22/taskops MCP server for Claude Code agents to manage persistent shared task boards with dependency graphs, atomic claims, and git-bound tasks. | |
| mcp-oidc4vci bernatmarcilla/mcp-oidc4vci Enables AI agents to inspect OIDC4VCI credential offers, discover issuer metadata, and orchestrate non-sensitive issuance protocol flows while keeping wallet-bound operations secure. |
33,751–33,800 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.