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-enterprise-spreadsheet-formula-dag-synthesizer-skill alphaparkinc/genpark-enterprise-spreadsheet-formula-dag-synthesizer-skill Enables agents to synthesize multi-tab enterprise spreadsheet financial models, detect circular references, compile formula dependency DAGs with functions like SUMIFS/XLOOKUP/CAGR/NPV, and evaluate deterministic preview cell matrices. | |
| genpark-ephemeral-cloud-resource-reaper-skill alphaparkinc/genpark-ephemeral-cloud-resource-reaper-skill Enables autonomous agents to detect and clean up idle or abandoned cloud GPU/disk resources for FinOps cost optimization. | |
| genpark-episodic-memory-consolidation-decay-skill alphaparkinc/genpark-episodic-memory-consolidation-decay-skill Enables AI agents to record episodic events, decay stale transient interactions using an Ebbinghaus forgetting curve, and consolidate frequent episodes into enduring semantic user persona traits for long-term personalization. | |
| genpark-episodic-semantic-memory-graph-synthesizer-skill alphaparkinc/genpark-episodic-semantic-memory-graph-synthesizer-skill Enables AI agents to synthesize and update hierarchical episodic and semantic user profile memory graphs through MCP, using deterministic zero-dependency Python processing with JSON Schema validation. | |
| genpark-event-driven-webhook-signature-verifier-skill alphaparkinc/genpark-event-driven-webhook-signature-verifier-skill Enables verifying event-driven webhook signatures and providing HMAC replay protection for AI agents and MCP clients. | |
| genpark-evm-flashloan-arbitrage-slippage-guard-skill alphaparkinc/genpark-evm-flashloan-arbitrage-slippage-guard-skill Provides autonomous EVM smart contract auditing and DeFi vulnerability triage, with a focus on flash loan arbitrage price impact and MEV sandwich protection for Aave and Uniswap v3. Enables deterministic callgraph verification and signature | |
| genpark-executive-meeting-action-item-matrix-extractor-skill alphaparkinc/genpark-executive-meeting-action-item-matrix-extractor-skill Extracts and structures executive meeting action items and decision accountability into JSON, enabling automated tracking and telemetry through MCP-compatible AI agents. | |
| genpark-financial-earnings-transcript-sentiment-tracker-skill alphaparkinc/genpark-financial-earnings-transcript-sentiment-tracker-skill Enables analysis of financial earnings call transcripts to track tone shifts and guidance sentiment using Finnhub data. | |
| genpark-fintech-transaction-velocity-fraud-sentinel-skill alphaparkinc/genpark-fintech-transaction-velocity-fraud-sentinel-skill Enables autonomous financial agents to submit payment webhooks, ledger entries, or transaction streams and receive balanced journal records, risk verification dossiers, and detection of card-testing clusters, volume spikes, and account take | |
| genpark-fintech-webhook-idempotent-replay-guard-skill alphaparkinc/genpark-fintech-webhook-idempotent-replay-guard-skill Enables agents to ingest payment webhooks and ledger transactions with cryptographic HMAC-SHA256 signature verification, sliding-window replay tolerance, and atomic idempotency locking. It also enforces double-entry debit/credit invariants | |
| genpark-flaky-test-timing-non-determinism-detector-skill alphaparkinc/genpark-flaky-test-timing-non-determinism-detector-skill Enables automated analysis of test suite traces to identify timing races and state pollution, helping detect flaky test non-determinism in CI/CD workflows such as Datadog CI or Trunk. | |
| genpark-fullstack-component-ast-patch-generator-skill alphaparkinc/genpark-fullstack-component-ast-patch-generator-skill Generates fullstack UI component AST patches with Tailwind tokens, enabling AI agents to produce structured UI patch outputs for frameworks like v0 and Lovable. | |
| genpark-generative-image-spatial-upscale-tiler-skill alphaparkinc/genpark-generative-image-spatial-upscale-tiler-skill Enables high-resolution image canvas tiling, spatial grid slicing, and seam blending for generative image upscaling workflows via MCP. | |
| genpark-graph-dijkstra-astar-pathfinder-skill alphaparkinc/genpark-graph-dijkstra-astar-pathfinder-skill Enables agents to compute shortest paths with Dijkstra and A*, resolve DAG dependencies, solve maximum flow, build minimum spanning trees, and rank graph centrality using JSON-RPC MCP tools. | |
| genpark-html-dom-semantic-tree-pruner-skill alphaparkinc/genpark-html-dom-semantic-tree-pruner-skill It enables autonomous browsing clients to canonically compress rendered HTML into compact semantic trees, filter navigation traps, bind webform schemas, transform tables into typed records, and score interaction trajectories. Its tools oper | |
| genpark-hubx-multimodal-mobile-agent-mcp alphaparkinc/genpark-hubx-multimodal-mobile-agent-mcp Enables autonomous agents to orchestrate HubX mobile AI tools for vision diagnostics, voice transcription, photorealistic image generation, and spatial design through a native MCP interface. | |
| Hume AI Empathic Voice Prosody & Expression Analyzer alphaparkinc/genpark-hume-ai-empathic-voice-prosody-expression-analyzer-skill Enables decoding of acoustic signals such as fundamental frequency, speaking rate, jitter, and pause ratio together with semantic sentiment to infer high-dimensional emotional states and emit real-time expressive TTS prosody directives. It | |
| genpark-hybrid-dense-sparse-bm25-rerank-fusion-skill alphaparkinc/genpark-hybrid-dense-sparse-bm25-rerank-fusion-skill Enables deterministic hybrid ranking by fusing dense, sparse, and BM25 retrieval results with reranking and Reciprocal Rank Fusion, producing structured output via MCP for AI agents. | |
| genpark-hybrid-rerank-fusion-retriever-skill alphaparkinc/genpark-hybrid-rerank-fusion-retriever-skill Enables fusing and reranking results from multiple search or retrieval sources using Reciprocal Rank Fusion and hybrid rank fusion. It provides deterministic scoring, robust parsing, and structured JSON output for autonomous agent workflows | |
| genpark-hypothesis-welch-t-test-statistical-evaluator-skill alphaparkinc/genpark-hypothesis-welch-t-test-statistical-evaluator-skill Enables MCP clients such as Claude Desktop, Cursor, and Windsurf to run two-sample Welch's t-tests that compute Satterthwaite degrees of freedom and two-tailed p-values for A/B test comparisons. It also exposes zero-dependency statistical p | |
| genpark-immutable-double-entry-ledger-kernel-skill alphaparkinc/genpark-immutable-double-entry-ledger-kernel-skill Enables autonomous financial agents and payment webhooks to create and verify immutable double-entry ledger entries, enforce debit-credit balance invariants, and generate hash-chained audit trails with fraud detection. | |
| genpark-influencer-brand-safety-controversy-auditor-skill alphaparkinc/genpark-influencer-brand-safety-controversy-auditor-skill Enables auditing influencer brand safety and multi-year controversy history through Traackr. | |
| genpark-influencer-share-of-voice-sentiment-synthesizer-skill alphaparkinc/genpark-influencer-share-of-voice-sentiment-synthesizer-skill Enables monitoring influencer brand share of voice and creator mention sentiment using Meltwater data, delivering deterministic structured output for MCP-compatible clients. | |
| InstacartCartSubstitutionAffinityResolver alphaparkinc/genpark-instacart-cart-substitution-affinity-resolver-skill Enables real-time out-of-stock grocery cart substitutions by enforcing dietary constraints, unit price delta ceilings, and brand affinity scoring to select suitable replacement products. | |
| genpark-instinct-proactive-intent-reflex-skill alphaparkinc/genpark-instinct-proactive-intent-reflex-skill Enables personal AI agents to sense ambient attention cues, compute intent probabilities, and execute reversible guarded dry-run actions without prompting. It routes routine reflexes through a fast System-1 decision layer backed by decaying | |
| genpark-int8-symmetric-per-tensor-quantizer-skill alphaparkinc/genpark-int8-symmetric-per-tensor-quantizer-skill Enables zero-dependency INT8 symmetric quantization of LLM weights and activations on a per-tensor or per-channel basis, reporting SNR and MSE reconstruction telemetry along with TTFT/TPOT latency and jitter metrics. Also exposes edge infer | |
| genpark-intent-signal-buying-propensity-scorer-skill alphaparkinc/genpark-intent-signal-buying-propensity-scorer-skill Enables predictive B2B outbound buying propensity scoring by aggregating time-decayed hiring, funding, and technographic signals, then classifying accounts into outbound action tiers. | |
| genpark-inverse-kinematics-joint-limit-solver-skill alphaparkinc/genpark-inverse-kinematics-joint-limit-solver-skill Enables deterministic inverse kinematics solving for 6-DOF robotic manipulators, converting poses to joint angles for UR5 and Franka Emika robots via MCP. | |
| genpark-jev-system1-subconscious-decision-skill alphaparkinc/genpark-jev-system1-subconscious-decision-skill Enables AI agents to resolve typed routing, loop-stall arbitration, and tool safety checks locally at sub-millisecond latency, avoiding unnecessary frontier LLM calls. It enforces reversible execution checkpoints, token budgets, and decayed | |
| genpark-jev-typed-state-action-router-mcp alphaparkinc/genpark-jev-typed-state-action-router-mcp Enables MCP-compatible agents to arbitrate typed state-action schemas, dispatch routine actions deterministically, and apply pre-escalation filters that reduce unnecessary frontier LLM calls. It supports low-latency, schema-enforced routing | |
| genpark-just-in-time-mcp-tool-compiler-skill alphaparkinc/genpark-just-in-time-mcp-tool-compiler-skill Enables agents to synthesize typed MCP tool schemas at runtime, validate Python ASTs for unsafe imports, compile sandboxed callable handlers, and hot-plug them into an active MCP runtime. It provides a zero-dependency, standard-library-only | |
| genpark-jwt-hmac-sha256-token-attestation-skill alphaparkinc/genpark-jwt-hmac-sha256-token-attestation-skill Enables agents to cryptographically verify JWT HMAC-SHA256 signatures, validate claims and enforce expiration, and perform Merkle inclusion proofs, Schnorr zero-knowledge identification, and nonce replay protection. Runs as a zero-dependenc | |
| genpark-knowledge-graph-entity-relation-triplet-extractor-skill alphaparkinc/genpark-knowledge-graph-entity-relation-triplet-extractor-skill Enables extraction of subject-predicate-object triples from text and multi-hop reasoning over connected data, producing Cypher and JSON-LD outputs for agent memory, vector search, and retrieval-augmented generation workflows. | |
| genpark-kv-cache-paged-attention-allocator-skill alphaparkinc/genpark-kv-cache-paged-attention-allocator-skill Enables MCP clients to manage LLM inference memory through virtual paged-attention KV-cache block mapping, non-contiguous physical page allocation, zero-copy fragmentation tracking, radix-trie prefix caching, INT8 quantized compute, and spe | |
| genpark-linear-regression-gradient-descent-engine-skill alphaparkinc/genpark-linear-regression-gradient-descent-engine-skill Enables agents to run zero-dependency statistical modeling and data analysis through MCP, including multivariate linear regression via gradient descent, anomaly detection, time-series forecasting, hypothesis testing, and PCA dimensionality | |
| genpark-linear-sprint-velocity-cycle-forecaster-skill alphaparkinc/genpark-linear-sprint-velocity-cycle-forecaster-skill Enables agile sprint backlog velocity calculation and milestone completion prediction through the Model Context Protocol, integrating with MCP-compliant clients for deterministic, low-latency forecasting. | |
| genpark-live-artifact-version-diff-tracker-skill alphaparkinc/genpark-live-artifact-version-diff-tracker-skill Enables developers and AI agents to compute deterministic in-memory unified diffs between successive versions of live artifacts and receive structured change telemetry in return. It runs with zero external dependencies and plugs into MCP-co | |
| genpark-long-horizon-autonomous-task-resumption-skill alphaparkinc/genpark-long-horizon-autonomous-task-resumption-skill Enables long-horizon autonomous task checkpointing and resumption through cryptographic event-sourced state persistence, idempotency key deduplication, timeout heartbeats, human-in-the-loop pause/resume tokens, and fail-safe recovery. | |
| genpark-low-latency-prosody-sentiment-modulator-skill alphaparkinc/genpark-low-latency-prosody-sentiment-modulator-skill Enables real-time voice agents to dynamically tailor speech pitch, cadence, and pause inflections through low-latency prosody, SSML, and emotion pacing modulation. | |
| genpark-managed-agent-fleet-lifecycle-supervisor-skill alphaparkinc/genpark-managed-agent-fleet-lifecycle-supervisor-skill Enables orchestration of worker registration, capability-based task dispatch, heartbeat telemetry, automatic circuit breaking, and checkpoint recovery for distributed agent operations. | |
| genpark-manus-autonomous-generalist-sandbox-executor-skill alphaparkinc/genpark-manus-autonomous-generalist-sandbox-executor-skill Provides a native MCP-compatible autonomous agent that decomposes general tasks, runs multi-modal web and sandbox execution loops, and self-corrects toward goal resolution using sensory and ambient inputs. | |
| Manus Autonomous Web Browser DOM Actuator alphaparkinc/genpark-manus-autonomous-web-browser-dom-actuator-skill This MCP server enables users to convert web page DOM structures into accessibility trees, filter non-interactive elements, generate robust CSS/XPath locators, and compile goal-driven multi-step browser automation plans with verification. | |
| genpark-markdown-table-to-json-transformer-skill alphaparkinc/genpark-markdown-table-to-json-transformer-skill Enables converting Markdown and HTML tables into structured JSON records with automatic column type inference and format normalization. It also offers supporting web-agent primitives such as DOM semantic pruning, form field auto-mapping, an | |
| genpark-maximum-flow-edmonds-karp-network-skill alphaparkinc/genpark-maximum-flow-edmonds-karp-network-skill Enables agents to compute shortest paths, topological dependency order, maximum flow and minimum-cut bottlenecks, minimum spanning trees, and PageRank centrality over weighted graphs for routing and execution planning. It exposes these grap | |
| genpark-merkle-tree-state-proof-verifier-skill alphaparkinc/genpark-merkle-tree-state-proof-verifier-skill Enables AI agents to construct SHA-256 Merkle trees and generate/verify inclusion proofs, validate JWT HMAC-SHA256 claims, run Schnorr zero-knowledge identification, and enforce constant-time secret comparison with nonce replay protection f | |
| genpark-meta-muse-episodic-memory-graph-mcp alphaparkinc/genpark-meta-muse-episodic-memory-graph-mcp Enables AI agents to combine ambient multimodal perception, continuous episodic memory, and recency-decay context briefing for fast deterministic decisions and safe tool routing. It integrates with MCP clients to reduce unnecessary frontier | |
| genpark-meta-muse-multimodal-episodic-resonance-mcp alphaparkinc/genpark-meta-muse-multimodal-episodic-resonance-mcp Enables personal AI agents to combine Meta Ray-Ban audio/vision sensory grounding with long-term episodic life memory graphs and empathetic emotional resonance. Integrates with MCP-compliant clients via JSON-RPC stdio for ambient, context-a | |
| genpark-milwaukee-onekey-smart-tool-inventory-skill alphaparkinc/genpark-milwaukee-onekey-smart-tool-inventory-skill Enables AI agents to track, calibrate, secure, and audit Milwaukee M18 FUEL ONE-KEY power tools through BLE proximity, geofencing, torque profiles, remote lockout, and fleet telematics. | |
| genpark-minimum-spanning-tree-kruskal-prim-skill alphaparkinc/genpark-minimum-spanning-tree-kruskal-prim-skill Enables MCP clients to compute shortest paths, topological dependency orderings, maximum network flows, minimum spanning trees, and PageRank-style centrality over graph topologies using only the Python standard library. It supports optimal | |
| genpark-monorepo-dependency-cycle-detector-skill alphaparkinc/genpark-monorepo-dependency-cycle-detector-skill Enables coding agents to analyze monorepo dependency graphs, detect and resolve circular dependencies, and plan a topological build order via Tarjan SCC and Kahn algorithms. It exposes deterministic, zero-dependency Python AST static analys |
27,451–27,500 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.