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 |
|---|---|
| reverse-information-paradox Diagnoses how consuming external AI forces you to reveal proprietary knowledge — your 'intelligence exhaust' (prompts, tool calls, corrections, evals) — to whoever controls the learning infrastructure, and prescribes a trust boundary that k | |
| RIA — ADV / Form CRS Compliance Checklist Activate when: annual ADV amendment season; onboarding/updating Form CRS delivery; launching marketing/testimonials; a mock/real exam; 'are we books-and-records compliant?', 'can we post this ad/testimonial?'. Do NOT activate when: the ques | |
| RIA — Client Behavior Coaching (Loss Aversion) Activate when: a market drop has clients wanting to sell; framing performance reviews; onboarding risk tolerance; 'clients panic in downturns,' 'how do I keep them invested?'. Do NOT activate when: a genuine plan change is warranted (don't | |
| RIA — Client Plan Margin of Safety Activate when: building a retirement/decumulation plan; setting cash buffers/withdrawal rates; stress-testing a plan; client near retirement; 'will they be okay if markets drop early?'. Do NOT activate when: pure accumulation with long hori | |
| RIA — Fiduciary Conflict-of-Interest Disclosure Activate when: identifying/disclosing conflicts in ADV 2A; compensation that varies by product; affiliated custodian/manager; rollover recommendations; 'is this a conflict we must disclose?'. Do NOT activate when: no advisory relationship / | |
| RIA — Position Sizing & Concentration Discipline Activate when: sizing a single-name or thematic position; a client with concentrated stock; setting max-position/rebalancing bands; 'how much of the portfolio should this be?'. Do NOT activate when: broad index allocation already governed b | |
| S-Curve Technology Adoption Activate when: user asks "why has our growth stalled after early success?", "when will this market saturate?", "we used to grow easily, now it's hard", "how do we cross the chasm?", "we need to reach mainstream buyers", "our marketing stopp | |
| scenario-planning Activate when: user says 'what are the scenarios,' 'stress test our strategy,' 'what if X happens,' 'multiple futures,' 'strategic resilience,' or is making a high-stakes irreversible decision with a 3+ year horizon where a single forecast | |
| The Second Curve Activate when: user says 'our growth is slowing and we need to figure out what's next', 'we keep doubling down on the core but I'm worried about disruption', 'when is the right time to start something new alongside the main business', 'shou | |
| Second-Order Thinking Activate when: user says 'and then what?', 'what are the second-order effects?', 'what could go wrong downstream?', 'what happens once everyone does this?', or brings a decision where the immediate effect is clear but downstream effects are | |
| Self-Evolution 3D Activate when: user says "I've been working hard but I'm not growing," development programs feel pointless, a leader has plateaued despite strong expertise, someone asks "how do I grow as a person / leader / founder," or high-quality output | |
| Self-Renewal Activate when: someone says "I keep reverting to old habits," "our strategy worked five years ago but not now," "I know I should think differently but can't seem to," "how do I stay relevant," or a leader/team is operating on prior-era assu | |
| The Seven Questions Test Activate when: user says 'seven questions,' 'Zero to One test,' 'Thiel checklist,' 'should I start this business,' 'is this startup idea any good,' or 'stress-test my startup idea'; a business idea, pivot, product line, or funding decision | |
| 势 — Strategic Momentum Model Activate when: user says 'we need to pick up speed', 'why aren't we getting traction', 'when should we scale / raise / launch', 'our efforts feel scattered', or 'are we ready to grow'; entering a new market or venture; post-pivot phase reas | |
| Signaling Games Activate when: user asks 'how do we prove we're high quality when they can't verify it', 'what credential or signal would work here', 'why is everyone spendi... | |
| situational-leadership Activate when: user says 'my management style isn't working for this person', 'I don't know how much to delegate', 'my top performer is disengaged', 'my new hire is struggling without guidance', 'should I give him more autonomy?', or asks h | |
| six-thinking-hats Activate when: a meeting keeps going in circles, people have different agendas, someone says 'we need structure' or 'let's look at this from all angles,' a proposal needs evaluation from multiple sides, or creative ideas keep getting killed | |
| social-proof Activate when: user says 'everyone is doing this,' 'join thousands of others,' 'it's the popular choice,' 'trusted by X customers,' wonders if a trend or consensus is real, suspects fake reviews or manufactured engagement, or is designing t | |
| SOP Systemization — Turn Repetitive Work into Repeatable Process Activate when: a founder is the bottleneck on repetitive work, or wants to delegate/automate to people or AI agents; 'everything depends on me', 'how do I document this', turning recurring tasks into repeatable processes. Do NOT activate wh | |
| status-quo-bias Activate when: user says 'we've always done it this way', 'changing now would be too disruptive', or 'no one is complaining so why change'; a team is slow to adopt a clearly better option; someone frames inaction as safe when omission carri | |
| strategic-commitment Activate when: someone says 'they don't believe we'll follow through', 'how do I make my threat credible?', 'how do we deter a competitor from entering?', 'we need to lock in this customer/partner', 'should we burn our bridges?', or when a | |
| Strategic Execution 3D9 Activate when: strategy results are consistently below target despite strong market conditions; a team agrees on a plan but results fall far short; diagnosing why execution broke down after a leadership change or merger; auditing whether th | |
| strategic-spiral-momentum Activate when: organization's strategy is losing momentum despite doing the right things; team disagrees on which strategic phase they're in; deciding whether to push deeper into current strategy or start a new cycle; setting phase-appropri | |
| strategy-execution-levers Activate when: 'our strategy is clear but nothing changes,' 'people understand the goal but revert to old behaviors,' 'initial launch energy faded after 2 months,' 'we keep having the same post-mortem findings,' or execution consistently fa | |
| structured-communication Activate when: user says 'they didn't get it,' 'I don't know how to bring this up,' 'the message wasn't clear,' 'how do I give feedback without sounding critical,' needs to structure a funding pitch or difficult conversation, or wants to di | |
| Sunk-Cost Fallacy Activate when: user says 'we've come too far to quit', 'we can't waste what we've already spent', 'should I keep going or cut my losses', 'we've invested too... | |
| survivorship-bias Activate when: user says 'look at what winners/billionaires/champions did,' investment returns or fund performance are being cited, a strategy is justified by pointing to companies that succeeded, historical data is treated as representativ | |
| switching-costs Activate when: user asks why customers don't switch to a better product, how to make a product stickier or build a moat, how to compete against an entrenched incumbent, why churn is low but revenue growth stalled, or mentions lock-in / vend | |
| SWOT Analysis Activate when: user says 'SWOT', 'our strengths and weaknesses', 'internal/external analysis', 'strategic audit', 'what's our edge', 'competitive threats we face', or is beginning a strategy planning cycle, evaluating market entry, or prepa | |
| T-Shaped Connector Activate when: someone says 'I'm T-shaped' and you want to verify it; a team can't translate insights across functions; a person asks what career move creates the most leverage; building a first startup team; a domain expert is hitting a ce | |
| Tax Prep — Filing-Position Consequence Cascade Activate when: a position/election has effects beyond this year — S-corp election, depreciation/§179/bonus, accounting-method change, filing status, Roth conversion, entity choice; client asks 'should I elect X?'. Do NOT activate when: the | |
| Tax Prep — Refundable-Credit Due-Diligence (Form 8867) Activate when: preparing any return claiming EITC, CTC/ACTC/ODC, AOTC, or Head-of-Household status; onboarding a paid-preparer workflow; a reviewer says 'did we document due diligence?', 'are we 8867-covered?', 'EITC audit risk.'. Do NOT ac | |
| Tax Prep — Pre-File Audit Premortem Activate when: a return has aggressive/uncertain positions (large Schedule C losses, high meals/vehicle, hobby-vs-business, real-estate professional status, R&D/ERC-type credits, crypto); before releasing a complex return; client says 'will | |
| Tax Prep — Season Throughput (Theory of Constraints) Activate when: returns are piling up before a deadline; reviewer/preparer is the bottleneck; 'we can't get through them all,' 'clients waiting on missing docs,' planning staffing for Jan–Apr. Do NOT activate when: off-season, low volume; th | |
| The Mom Test — Customer Interviews That Don't Lie to You Activate when: a founder is about to talk to customers/users to validate an idea; 'I asked people and they loved it', 'how do I run customer interviews', 'is there demand for this?', pre-PMF discovery. Do NOT activate when: the question is | |
| Theory of Constraints Activate when: user says 'everyone is working hard but results are flat', 'where is our bottleneck', 'we keep adding capacity but throughput doesn't improve', 'backlog piling up at one stage', 'Goldratt / TOC / Five Focusing Steps', or is d | |
| thought-experiment Activate when: user says 'what if,' 'let's reason through this,' 'imagine that,' or 'thought experiment'; a strategic decision is too risky or irreversible to test empirically; someone wants to refute a claim by reasoning rather than data; | |
| 1,000 True Fans Activate when: a creator/SMB/indie founder needs a sustainable audience-based living without going viral; 'do I need to be famous to make this work', building direct audience revenue; niche monetization. Do NOT activate when: the model genu | |
| Three-Radius Model — 认知/能力/行动三半径 Activate when: user says 'I know what to do but I'm not doing it', 'we understand the problem but can't fix it', 'effort isn't translating to results', 'why isn't our team executing?', or 'I feel stuck despite being competent'. Also activat | |
| Thucydides's Trap Activate when: user asks about US-China war risk or great-power rivalry; someone says 'can they avoid conflict' or 'is escalation inevitable'; analyzing geopolitical risk in supply chains, FDI, or market entry; a new market entrant threaten | |
| Timing-Action Matrix Activate when: a major decision is imminent and the team hasn't explicitly checked whether NOW is the right time; a correct strategy keeps hitting unexpected resistance; a post-mortem asks why something failed despite being the right call; | |
| tipping-point Activate when: user asks 'how close are we to critical mass', 'why did growth suddenly explode (or collapse)', 'will this trend keep spreading', 'is there a network effect threshold here', 'what if we concentrated effort on early adopters'. | |
| Tragedy of the Commons Activate when: shared resource is degrading even though no one is doing anything "wrong"; team complains about free riders or overuse; designing API quotas, rate limits, or open-source governance; someone says "everyone is overusing this" o | |
| Travel Advisor — Cancellation & Refund Decision Activate when: a client needs to cancel/change; deciding refund vs credit vs insurance claim; comparing supplier penalties; 'what's my best move to recover the most?'. Do NOT activate when: fully refundable booking, no decision needed. More | |
| Travel Advisor — Commission & Markup Pricing Activate when: setting planning fees vs commission; deciding markup on net rates; 'should I charge a fee or rely on commission?'; pricing a service tier; low-commission or non-commissionable components. Do NOT activate when: fixed franchise | |
| Travel Advisor — Peak-End Experience Design Activate when: designing an itinerary you want remembered and referred; planning honeymoon/milestone/luxury trips; 'how do I get repeat clients and referrals?'; setting surprise-and-delight touches. Do NOT activate when: pure transactional | |
| Travel Advisor — Repeat-Client Relationship Ladder Activate when: turning one-trip bookers into annual clients; building a client-for-life book; planning post-trip nurture; 'how do I stop starting from zero every year?'. Do NOT activate when: pure one-off transactional volume with no repeat | |
| Travel Advisor — Trip Contingency Planning Activate when: building a complex/high-value itinerary; multi-supplier or international trip; hurricane/advisory season; client asks 'what if something goes wrong?'; designing a group/FIT trip. Do NOT activate when: single refundable bookin | |
| trust-brand-architecture Activate when: brand perception lags product quality; a values campaign generated backlash ('they say X but do Y'); NPS and brand sentiment diverge; marketing underperforms despite a good product; someone asks 'how do we build brand trust?' | |
| Tunnel Vision & Slack Activate when: user says 'I'm always firefighting,' 'I never have time to think,' 'everything feels urgent,' 'I can't get ahead,' or 'we keep pushing strategic work because of day-to-day fires.' Do NOT activate when: the urgency is a single |
24,951–25,000 of 108,824
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.