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 |
|---|---|
| genpark-agent-multimodal-document-table-extractor-skill alphaparkinc/genpark-agent-multimodal-document-table-extractor-skill Enables AI agents and MCP clients to convert ASCII/OCR text dumps, aligned spacing, and CSV/TSV data into clean GitHub Flavored Markdown tables and typed JSON deterministically. | |
| PRD & Technical Spec Synthesizer alphaparkinc/genpark-agent-prd-spec-synthesizer-skill Enables autonomous agents to turn high-level features into structured PRDs, user stories, Gherkin acceptance criteria, edge cases, and OpenAPI-compatible API contract skeletons. It exposes MCP tools for synthesizing requirements and generat | |
| Prompt Guard Injection Sanitizer alphaparkinc/genpark-agent-prompt-guard-injection-sanitizer-skill Enables agents to scan untrusted text for jailbreak markers, prompt-override exploits, and indirect injection payloads, then cleanse it by replacing hostile command sequences with safe placeholders before the text reaches a prompt. Exposes | |
| genpark-agent-prompt-injection-jailbreak-sentinel-skill alphaparkinc/genpark-agent-prompt-injection-jailbreak-sentinel-skill Enables MCP clients and autonomous agents to screen incoming prompts for system overrides, delimiter breakouts, and encoded payload evasion, helping block prompt injection and jailbreak attempts before execution. | |
| genpark-agent-prompt-injection-sanitizer-firewall-skill alphaparkinc/genpark-agent-prompt-injection-sanitizer-firewall-skill Enables autonomous agents and MCP-compatible clients to route inputs through a deterministic firewall that flags and sanitizes prompt injection, jailbreaks, system prompt exfiltration, and tool hijacking. It supports JSON-RPC stdio integrat | |
| genpark-agent-rate-limit-token-bucket-regulator-skill alphaparkinc/genpark-agent-rate-limit-token-bucket-regulator-skill Enables AI agents and multi-agent frameworks to throttle outbound LLM API calls with multi-provider token-bucket and sliding-window limiting, plus predictive token-consumption budgeting to avoid TPM/RPM throttling. It runs deterministically | |
| genpark-agent-recursive-self-critique-reflection-skill alphaparkinc/genpark-agent-recursive-self-critique-reflection-skill Enables MCP-compliant agents to run deterministic iterative self-critique and reflection loops that autonomously evaluate and revise outputs to repair logic defects, test failures, and hallucinations. It connects via JSON-RPC stdio with no | |
| GenParkAgentRedTeamIndirectInjectionFuzzer alphaparkinc/genpark-agent-redteam-indirect-injection-fuzzer-skill Enables red-team testing of AI agents against indirect prompt injection by fuzzing payloads with delimiter evasion and Unicode homoglyphs, detecting exfiltration canaries, enforcing least-privilege tool boundaries, and producing cryptograph | |
| genpark-agent-runtime-self-healing-diagnostics-skill alphaparkinc/genpark-agent-runtime-self-healing-diagnostics-skill Enables autonomous agents to detect configuration drift, corrupted execution state, and missing runtime resources, then perform root-cause classification and automated snapshot rollback for self-healing runtime recovery. | |
| Sales Pitch & Objection Handler Skill alphaparkinc/genpark-agent-sales-pitch-objection-handler-skill Enables autonomous enterprise sales agents to generate customized B2B proposals, ROI framing, competitive battlecard talking points, and tailored objection-handling scripts through the generate_sales_pitch MCP tool. | |
| genpark-agent-sandboxed-virtual-code-runner-skill alphaparkinc/genpark-agent-sandboxed-virtual-code-runner-skill Enables agents and MCP clients to safely execute and verify multi-language code in a deterministic sandbox, using AST security analysis to block dangerous system calls and capture isolated execution telemetry. | |
| genpark-agent-self-consistency-consensus-scorer-skill alphaparkinc/genpark-agent-self-consistency-consensus-scorer-skill Enables aggregating divergent LLM reasoning trajectories via multi-path chain-of-thought self-consistency consensus scoring and majority voting into high-confidence deterministic outputs. Plugs into any MCP-compliant client over JSON-RPC 2. | |
| Semantic Cache Query Matcher alphaparkinc/genpark-agent-semantic-cache-query-matcher-skill Enables agents to store and retrieve LLM query-response pairs using fuzzy cosine similarity matching with configurable thresholds and cache hit/miss telemetry. This reduces inference cost and latency by reusing semantically similar cached r | |
| genpark-agent-semantic-cache-similarity-deduplicator-skill alphaparkinc/genpark-agent-semantic-cache-similarity-deduplicator-skill Enables high-throughput semantic query caching and approximate deduplication by combining exact SHA-256 matching, cosine similarity threshold checks, and LRU eviction. | |
| genpark-agent-session-compaction-tokenizer-skill alphaparkinc/genpark-agent-session-compaction-tokenizer-skill Enables AI agents and developers to dynamically tokenize session context windows and compact dialogue history into structured outputs via MCP, CLI, or Python client. Runs on pure standard library Python with no external dependencies for det | |
| Slide Layout Auto-Rebalancer Skill alphaparkinc/genpark-agent-slide-layout-auto-rebalancer-skill Enables agents to optimize presentation layouts by adjusting element coordinates and bounding boxes, eliminating overlaps and canvas overflow, and harmonizing typographic spacing. | |
| Slide Theme & Palette Harmonizer MCP Server alphaparkinc/genpark-agent-slide-theme-palette-harmonizer-skill Enables agents to generate harmonized presentation design themes from a primary brand color, including WCAG AA/AAA contrast checks, typography pairings, and semantic styling tokens. | |
| genpark-agent-speculative-decoding-orchestrator-skill alphaparkinc/genpark-agent-speculative-decoding-orchestrator-skill Enables agents to orchestrate speculative decoding runs and deterministically verify draft-model acceptance probabilities via MCP. | |
| genpark-agent-synthetic-data-differential-privacy-guard-skill alphaparkinc/genpark-agent-synthetic-data-differential-privacy-guard-skill Enables safe agentic execution and privacy-preserving data release by generating epsilon-noise and redacting quasi-identifiers in tabular datasets, while guarding against prompt injection, runaway spend, planning loops, and destructive shel | |
| genpark-agent-synthetic-trajectory-evaluator-skill alphaparkinc/genpark-agent-synthetic-trajectory-evaluator-skill Enables autonomous web agents to prune DOM trees, canonicalize URLs against anti-crawler traps, auto-map form input schemas, and convert HTML/Markdown tables into typed JSON records. It also scores agent execution trajectories against gold | |
| genpark-agent-tool-call-self-healing-retry-guard-skill alphaparkinc/genpark-agent-tool-call-self-healing-retry-guard-skill Enables autonomous agents to automatically repair and guard tool-call schemas through deterministic AST analysis and parameter projection validation, ensuring reliable, schema-compliant interactions in MCP-compatible environments. | |
| genpark-agent-tool-dag-speculative-prefetcher-skill alphaparkinc/genpark-agent-tool-dag-speculative-prefetcher-skill Enables agents and MCP-compliant clients to resolve tool execution graphs by scheduling calls as a topological dependency DAG, running independent tools in parallel batches while speculatively prefetching and evaluating arguments. This cuts | |
| genpark-agent-tool-retry-exponential-backoff-jitter-skill alphaparkinc/genpark-agent-tool-retry-exponential-backoff-jitter-skill This MCP server wraps agent tool calls with fault-tolerant retry logic, including adaptive exponential backoff, full jitter, 429 rate-limit backpressure handling, and circuit breaker patterns. It enables autonomous agents to execute tools r | |
| genpark-agent-web-stealth-scraper-distiller-skill alphaparkinc/genpark-agent-web-stealth-scraper-distiller-skill Enables MCP-compatible agents to scrape web pages and distill them into clean Markdown by removing cookie banners, navbars, and ads for efficient consumption. | |
| Agentcard MCP Server alphaparkinc/genpark-agentcard-autonomous-payment-skill Enables AI agents to autonomously issue single-use, merchant-locked virtual Visa cards and execute payments across supported merchants through the Model Context Protocol, with human-in-the-loop approval. | |
| genpark-agentic-cache-semantic-deduplicator-skill alphaparkinc/genpark-agentic-cache-semantic-deduplicator-skill Enables agents to deduplicate semantic prompt caches by measuring Jaccard token overlap, preventing redundant LLM calls and reducing cost. It provides deterministic lookup-or-insert cache behavior through an MCP-compatible interface. | |
| AgenticCartAbandonmentRecoveryOrchestrator alphaparkinc/genpark-agentic-cart-abandonment-recovery-orchestrator-skill Enables users to diagnose cart abandonment friction, generate margin-preserving incentive vouchers, and synthesize channel-native win-back messages across commerce platforms. It supports recovery orchestration for abandoned checkout session | |
| Agentic Cold Email Outreach Spam Score & Deliverability Guard alphaparkinc/genpark-agentic-cold-email-outreach-spam-score-guard-skill Enables AI agents to audit B2B cold email copy for spam triggers and inbox placement risk by scoring subject lines and bodies against SpamAssassin keyword heuristics, exclamation and capitalization ratios, unsubscribe compliance, and readin | |
| genpark-agentic-commerce-a2m-bargaining-negotiator-skill alphaparkinc/genpark-agentic-commerce-a2m-bargaining-negotiator-skill Enables AI agents to run automated Rubinstein alternating-offers bargaining and multi-item bundling negotiations with merchant gateways over MCP STDIO JSON-RPC. Supports game-theoretic surplus splitting and agentic commerce negotiation work | |
| genpark-agentic-commerce-agent-gateway-verifier-skill alphaparkinc/genpark-agentic-commerce-agent-gateway-verifier-skill This MCP server enables agents to authenticate legitimate consumer shopping agents through cryptographic proof-of-delegation and block scraper abuse at retail gateways. It also exposes machine-to-machine catalog resolution, spend-bounded to | |
| genpark-agentic-commerce-ap2-payment-delegation-vault-skill alphaparkinc/genpark-agentic-commerce-ap2-payment-delegation-vault-skill Enables AI agents to manage ephemeral payment tokens with cryptographic spend mandates, risk scoring, catalog resolution, and automated bargaining through native MCP tools. | |
| AI Agentic Commerce Review Hijacking & Authenticity Sentinel alphaparkinc/genpark-agentic-commerce-counterfeit-product-review-synthesizer-skill Enables auditing product review streams for review-velocity clustering, sentiment dissonance, repetitive incentivized phrases, and review hijacking. Produces adjusted authentic ratings and authenticity indexes for e-commerce products. | |
| Agentic Escrow Milestone Payout Arbiter alphaparkinc/genpark-agentic-commerce-dynamic-escrow-milestone-releaser-skill Enables autonomous verification of contractual milestones—such as merged pull requests, test pass rates, and Lighthouse scores—against cryptographic acceptance proofs to authorize escrow fund releases. | |
| genpark-agentic-commerce-intent-cart-checkout-skill alphaparkinc/genpark-agentic-commerce-intent-cart-checkout-skill Enables agents to translate natural-language shopping intent into aggregated multi-merchant item discovery, consolidated cross-merchant carts, and structured transaction checkout payloads. Runs as a zero-dependency, MCP-compatible engine th | |
| Agentic Commerce Price Drop Retroactive Refund Claimer alphaparkinc/genpark-agentic-commerce-price-drop-retroactive-refund-claimer-skill Enables scanning past order receipts against retailer price-protection policy windows, verifying eligibility, and synthesizing auto-claim refund filing packets for retailers such as Target, Best Buy, Amazon, and Nordstrom. | |
| AI Agentic Commerce Telecom & Utility Bill Retention Negotiator alphaparkinc/genpark-agentic-commerce-recurring-utility-bill-negotiator-skill Enables users to audit monthly broadband and mobile bills, detect out-of-contract price creep and hidden junk fees, and generate high-leverage churn-threat retention scripts. | |
| Agentic Commerce RFQ Multi-Vendor Bidding Arbiter alphaparkinc/genpark-agentic-commerce-rfq-multi-vendor-bidding-arbiter-skill Enables agents to submit wholesale quotations from competing suppliers and receive a ranked, deterministic arbitration outcome scoring unit volume pricing, minimum order quantities, landed tariffs, net payment terms, and historical lead-tim | |
| GenPark Agentic Commerce Sponsored Placement Router alphaparkinc/genpark-agentic-commerce-sponsored-placement-router-skill Enables AI agents to resolve multi-merchant catalogs, negotiate prices via automated alternating-offer bargaining, and execute spend-bounded tokenized payments over UCP and AP2 agentic commerce rails. It also optimizes in-context ad slot al | |
| genpark-agentic-commerce-ucp-catalog-resolver-skill alphaparkinc/genpark-agentic-commerce-ucp-catalog-resolver-skill Enables AI agents to resolve multi-merchant Universal Commerce Protocol catalogs with Pareto-optimal pricing and SLA-aware routing, while supporting tokenized payment delegation and automated bargaining. | |
| Agentic Contextual Prompt Compressor alphaparkinc/genpark-agentic-contextual-compression-prompt-packer-skill This MCP server enables clients to reduce oversized prompts by selecting task-relevant context, removing lexical duplicates, and pruning to a token budget while preserving critical system invariants. | |
| AgenticFreightCarbonOffsetRouter alphaparkinc/genpark-agentic-cross-carrier-freight-carbon-offset-router-skill Enables users to compare and select multi-carrier freight routes by applying GLEC emissions factors across air, rail, maritime, and electric ground modes while weighing transit time and landed cost. It also calculates the resulting CO2 foot | |
| genpark-agentic-delegated-escrow-settlement-ledger-skill alphaparkinc/genpark-agentic-delegated-escrow-settlement-ledger-skill Enables multi-party conditional escrow locking, autonomous proof-of-delivery release triggers, referral fee splits, and dispute freeze protection for agentic commerce and enterprise settlement workflows. | |
| genpark-agentic-hallucination-grounding-checker-skill alphaparkinc/genpark-agentic-hallucination-grounding-checker-skill Enables sentence-level factual grounding verification and hallucination detection for RAG outputs via MCP, with deterministic scoring and structured telemetry. | |
| Agentic Inventory Restock Predictive Balancer alphaparkinc/genpark-agentic-inventory-restock-predictive-balancer-skill Enables agents to forecast SKU sales velocity, calculate safety stock and reorder points, and autonomously generate supplier purchase orders when stock is running low. It exposes this supply-chain analysis over JSON-RPC 2.0 tools that can b | |
| AgenticK8sSpotTerminationDrainScheduler alphaparkinc/genpark-agentic-kubernetes-node-spot-termination-drain-scheduler-skill Enables users to intercept cloud spot termination notices and coordinate Kubernetes drain operations by issuing cordon commands and prioritized, graceful pod eviction sequences so critical workloads can be redistributed with minimal downtim | |
| genpark-agentic-lead-score-velocity-evaluator-skill alphaparkinc/genpark-agentic-lead-score-velocity-evaluator-skill Enables autonomous agents to score customer leads and evaluate interaction velocity to predict conversions, exposing deterministic, zero-dependency evaluation as an MCP tool. Integrates with MCP clients like Claude Desktop and Cursor for pr | |
| AgenticMultimodalInvoiceReceiptReconciler alphaparkinc/genpark-agentic-multimodal-invoice-receipt-reconciler-skill Enables parsing of structured receipts, invoices, and bank statements and tolerance-based three-way reconciliation that detects tax calculation drift and duplicate claims, then synthesizes balanced double-entry bookkeeping journal entries. | |
| genpark-agentic-multimodal-pdf-contract-redline-diff-skill alphaparkinc/genpark-agentic-multimodal-pdf-contract-redline-diff-skill Enables comparison of baseline and counter-party contract clause revisions, flags high-risk shifts in indemnity, IP assignment, liability caps, and non-solicitation, and generates legal redline summaries with severity and escalation verdict | |
| genpark-agentic-multimodal-product-return-fraud-detector-skill alphaparkinc/genpark-agentic-multimodal-product-return-fraud-detector-skill Enables detection and risk scoring of e-commerce return fraud and wardrobing by correlating return velocity, carrier scan discrepancies, product tag tampering indicators, and customer return history, then enforcing actions like restocking f | |
| genpark-agentic-negotiation-counter-offer-strategist-skill alphaparkinc/genpark-agentic-negotiation-counter-offer-strategist-skill Enables autonomous multi-turn negotiation by calculating the Zone of Possible Agreement, modeling concession decay curves, enforcing reservation price thresholds, and generating structured bargaining counter-offers. It supports agentic comm |
27,301–27,350 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.