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 |
|---|---|
| Chief Dreamer A private, persistent memory and method for your dreams and ideas. Sign-in required (OAuth). | |
| Domain Health Domain checks: SPF, DKIM, DMARC, BIMI, MTA-STS, SSL/TLS and HTTP security headers. | |
| Dry Run Monte Carlo cash-flow simulator for a small services business: odds of running short, and why. | |
| Small Business Funding Finder Searches live federal grant listings on Grants.gov and reads full opportunity details. | |
| Idea to Product Sequencer Rule-based product method: scores an idea, finds its riskiest assumption, sequences the build. | |
| Own It Maps who controls a business web presence (domain, DNS, hosting, site) and the cost of leaving. | |
| Owner's Desk An owner's own Stripe: who owes you, today's money, draft invoices. Every write needs a confirm. | |
| Package Pricing Lab Builds fixed-price service packages from costs and capacity, with margin and capacity math. | |
| Site Check Public website health scan, AI-crawler and structured-data readiness, and a local presence score. | |
| Local Trades Kit US data for home-service trades: service-area ZIPs, seasonal demand, review timing, service pages. | |
| Voice Agent Linter Lints a phone voice agent prompt and platform config before real callers reach it. | |
| Mode VC Chart: AI Adoption Rate Among US Businesses 23.8% of US firms used AI in the latest two-week survey, up from 17.3% 10 months earlier.. | |
| Mode VC Chart: AI Chatbot Adoption by Age: ChatGPT Hits 61% in Youth 61% of Americans aged 18–29 use ChatGPT; Claude reaches 6% of all adults. Query full empirical. | |
| Mode VC Chart: AI Coding Assistant Statistics: Use vs. Trust 66% of developers use AI coding assistants, but only 48% trust AI output they can validate, a. | |
| Mode VC Chart: AI in the Workplace Statistics: What Workers Gain Among US employees who use AI, 63% say it makes them faster but only 46% say it raises. | |
| Mode VC Chart: AI Inference Cost: Output Token Pricing for Flagship LLMs Qwen3.8-Max rates within 30 Arena points of Claude Opus 5.5 for 3.3× less per output token.. | |
| Mode VC Chart: AI Model Launch Strategy: What Labs Ship on Day One Of 13 model launches from 23 Sept 2026 to 3 Oct 2026, 9 opened an API on day one, 7 of them. | |
| Mode VC Chart: AI Model Releases in 2026: Which Labs Ship Most 14 labs launched 50 named AI models from 6 Jul 2026 to 27 Sept 2026. Alibaba's Qwen team led. | |
| Mode VC Chart: AI Productivity Statistics: What AI Users Say They Gain 86% of surveyed Claude users say AI makes them faster and 82% say it widens what they can do.. | |
| Mode VC Chart: Anthropic SDK vs. OpenAI SDK: npm Downloads Anthropic's SDK went from 34% to 97% of the OpenAI SDK's weekly npm downloads in 46 weeks. It. | |
| Mode VC Chart: Anthropic vs. OpenAI vs. Google: Who Holds the Top 20? Anthropic holds 8 of the Arena top 20, Google 7 and OpenAI 0. In Apr 2025 OpenAI held 9 and. | |
| Mode VC Chart: Arena Rating Gap: Google Leads; Anthropic Trails by 22 Pts Google leads the Arena again; Anthropic trails by 22 points and OpenAI by 62. Query full. | |
| Mode VC Chart: Best AI for Creative Writing: Arena Rankings Anthropic has 7 of the top 10 creative writing models and Google 3. Meta's muse-spark-1.2. | |
| Mode VC Chart: Best AI for Math: The Arena Math Leaderboard 14 models are statistically tied for #1 in Arena math, spanning 49 points. 4 are Flash models. | |
| Mode VC Chart: Best AI Model vs. Most Popular AI Chatbot Anthropic builds 14 of the Arena's top 50 models and OpenAI builds 5. But 44% of US adults use. | |
| Mode VC Chart: Best AI Models by Category: Google vs. Anthropic Ranks Google holds #1 in 6 of 8 Arena categories; Anthropic in 2. OpenAI has no category in the top. | |
| Mode VC Chart: Best LLM for Coding: The Arena Coding Leaderboard claude-opus-4-6-high leads Arena coding at 1535. Anthropic has held #1 for 280 days, but Z.ai. | |
| Mode VC Chart: Which AI Labs Get the Most Chatbot Arena Votes? OpenAI's models hold 17.9% of all 12.3M Arena votes and none of the top 20 seats. Anthropic's. | |
| Mode VC Chart: ChatGPT vs. Gemini vs. Meta AI: User Numbers Compared ChatGPT reports 900M weekly users (Feb 2026). Gemini reports 950M monthly (Jul 2026), 2.4× its. | |
| Mode VC Chart: Chinese AI Models vs. US Models: How Close Is China? qwen3.8-max trails claude-fable-5.1-max by 27 points. Over the past year the gap has stayed. | |
| Mode VC Chart: Claude vs ChatGPT: Which Ranks Higher on the Leaderboard? On 11 Sept 2026, Claude's best model rates 40 points above OpenAI's and 16 above Gemini's.. | |
| Mode VC Chart: Gemini Leaderboard Ranking: Google's Days at #1 Since Jan 2024, Google's models have been #1 on Arena for 227 of 1003 days, third among labs. | |
| Mode VC Chart: Generative AI Usage Statistics for the US 54.8% of Americans aged 18–64 used generative AI outside work in Q2 2026, up from 36.0% in Q3. | |
| Mode VC Chart: How Long Does the #1 AI Model Stay on Top? Since Jan 2025, Arena's #1 spot has changed hands 16 times. Half of the finished reigns lasted. | |
| Mode VC Chart: How Many AI Models Are There? Arena Rates 399 Arena rates 399 text models from 52 labs, up from 10 models in May 2023. Yet only 5 labs hold. | |
| Mode VC Chart: LangChain Alternatives: Downloads vs. Newer AI SDKs LangChain's share of weekly npm framework downloads fell from 32% to 19% in 46 weeks, while. | |
| Mode VC Chart: LLM API Pricing History: Cost per Million Tokens OpenAI's headline model fell from $30 to $2 per million input tokens since Jun 2023: 93% in. | |
| Mode VC Chart: LLM Leaderboard Top 10: #1 Leads #10 by 18 Points The gap between Arena's #1 and #10 models fell from 148 points in Feb 2024 to 18. #10 now wins. | |
| Mode VC Chart: LLM Rankings by Category: The Hidden Specialists claude-opus-4-5-20251101-high-32k is #49 overall but #14 in coding. gpt-5.6-terra-xhigh is #52. | |
| Mode VC Chart: MCP Adoption: Model Context Protocol SDK Downloads The MCP SDK had 52.3M npm downloads in the week of 24 Aug, 1.3× the OpenAI SDK and more than. | |
| Mode VC Chart: Ollama Downloads: Is Local LLM Adoption Keeping Up? On npm, Ollama's client fell from 27 to 6.6 downloads per 1,000 cloud lab SDK downloads in 46. | |
| Mode VC Chart: Open-Source vs. Closed AI Models: How Big Is the Gap? The best open model trails the best closed model by 42 Arena points. In a blind vote it would. | |
| Mode VC Chart: Open-Weight Models in 2026: How Many Launch Open? Of 51 named model launches since 6 Jul 2026, 22 stated their weights: 18 closed, 4 open. All. | |
| Mode VC Chart: Open Weights vs. Proprietary AI: The Gap Stays at 42 Points Open weights briefly beat every proprietary model in early 2025 (deepseek-r1); today the best. | |
| Mode VC Chart: OpenAI Leaderboard Ranking: 8th Among Labs on Arena OpenAI was the #1 lab on Arena for 95% of 2024. On 25 Sept 2026 it ranks 8th among labs, and. | |
| Mode VC Chart: OpenAI Python Package Downloads: Who Grew Since August Between 31 Aug–6 Sep and 25 Sep–1 Oct, anthropic's PyPI downloads grew 25% and openai's 14%,. | |
| Mode VC Chart: Pre-Seed Valuation Dispersion 2026: Capital Concentration at the Top Decile The top 10% of pre-seed AI startups command 3.4x higher valuations ($28.0M vs $8.2M) than. | |
| Mode VC Chart: Vercel AI SDK Downloads vs. the Lab SDKs The Vercel AI SDK fell from 31 to 23 weekly downloads per 100 downloads of the OpenAI,. | |
| Mode VC Chart: Voice AI Models in 2026: Launches by Modality Audio and speech made 11 of 59 named model launches since 6 Jul 2026, from 6 labs. In the. | |
| Mode VC Atlas: 8VC Partner & Fund Intelligence Query partner roster, headquarters (California), sector focus, and fund benchmarks for 8VC in. |
33,101–33,150 of 50,909
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.