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 |
|---|---|
| snip-mcp alparse/snip-mcp An MCP server for capturing screen regions directly into Claude Code conversations via hotkeys or tool calls. It provides Windows-specific screen snipping with multi-monitor support, persistent storage, and automatic management of captured | |
| ÖPNV MCP Server alpenkind/oepnv-mcp MCP server for querying German public transport information, including station search, live departures/arrivals, and journey planning. | |
| appnova-app-store-connect-mcp alperduzgun/appnova-app-store-connect-mcp Enables managing your iOS app's entire lifecycle with natural language through App Store Connect, offering 48 tools across 14 categories for ASO, reviews, analytics, subscriptions, pricing, and more. | |
| Call For Papers MCP alperenkocyigit/call-for-papers-mcp Enables searching for upcoming academic conferences and events from WikiCFP by keywords, returning detailed information including dates, locations, submission deadlines, and related resources. | |
| Semantic Scholar MCP Server alperenkocyigit/semantic-scholar-graph-api A comprehensive Model Context Protocol server that provides AI assistants with direct access to Semantic Scholar's academic database, enabling advanced paper discovery, citation analysis, author research, and AI-powered recommendations. | |
| VSCode MCP Python Server alpermdr/vscode_mcp_python A production-ready MCP server that enables Claude to interact with local codebases through file operations, code analysis, and VSCode integration tools. It optimizes performance using LRU caching and asynchronous execution while maintaining | |
| HackerOne MCP Server alpernae/hackerone-mcp Enables MCP clients like Claude and Codex to interact with HackerOne's API to list and get reports, programs, and scopes. | |
| Intigriti MCP Server alpernae/intigriti-mcp A Model Context Protocol (MCP) server for interacting with the Intigriti bug bounty platform's Researcher API. It enables AI assistants to manage bug bounty programs, submissions, and research workflow. | |
| ssh-admin alpertk/ssh-admin MCP server for managing SSH connections to remote servers, providing host registry, session pool, and secure command execution with read-only protection. | |
| genpark-actuator-thermal-torque-throttle-sentinel-skill alpha-park/genpark-actuator-thermal-torque-throttle-sentinel-skill Monitors robotic joint actuator thermal and torque loads, triggering preventative throttling for DJI/Robomaster hardware and enabling real-time actuator safety supervision through MCP. | |
| genpark-adaptive-token-rate-limiter-skill alpha-park/genpark-adaptive-token-rate-limiter-skill Enables adaptive token bucket rate limiting and backoff scheduling to prevent HTTP 429 errors for autonomous agents and API clients. Provides a deterministic zero-dependency MCP/CLI engine for acquiring token permits and returning structure | |
| Adaptive Retry Circuit Breaker MCP Server alpha-park/genpark-agent-adaptive-retry-circuit-breaker-skill Enables agents to guard external API calls with circuit-breaker health checks, success/failure reporting, and recovery state inspection, including exponential backoff with jitter and fail-fast rejection. | |
| genpark-agent-bash-command-sandbox-guard-skill alpha-park/genpark-agent-bash-command-sandbox-guard-skill Enables autonomous agents to run shell and subprocess commands through a guardrail that blocks destructive system mutations, path traversals, and reverse shells. It also adds prompt-injection detection, spend circuit breakers, loop detectio | |
| genpark-agent-budget-token-spend-circuit-breaker-skill alpha-park/genpark-agent-budget-token-spend-circuit-breaker-skill Enables agents to track real-time token spend and cost velocity, automatically throttle or freeze execution when budgets are exceeded, and enforce sandbox and safety guardrails. | |
| genpark-agent-codebase-ast-dependency-slicer-skill alpha-park/genpark-agent-codebase-ast-dependency-slicer-skill Enables coding agents and MCP-compatible clients to parse multi-file AST dependency graphs and extract minimum context through symbol call-tree slicing. It provides deterministic, zero-dependency execution over JSON-RPC stdio. | |
| Conflict Resolution Merger Skill for Autonomous Agents alpha-park/genpark-agent-conflict-resolution-merger-skill Enables multiple autonomous agents to safely modify shared workspaces and documents by performing three-way reconciliation, detecting collisions, and returning structured conflict data for automated re-arbitration. | |
| genpark-agent-context-dynamic-compressor-skill alpha-park/genpark-agent-context-dynamic-compressor-skill Enables MCP-compatible clients and autonomous agents to dynamically compress long-horizon context by pruning verbose tool outputs, HTML boilerplate, and semantic redundancy by 40-70%. It reduces token burn and latency through deterministic, | |
| Continuous Context Session Compactor alpha-park/genpark-agent-continuous-context-session-compactor-skill Enables autonomous agents to keep long-running conversations within configurable token budgets by compacting message arrays, preserving system instructions and recent turns while distilling older history into concise summaries. This prevent | |
| genpark-agent-cryptographic-audit-trail-merkle-sealer-skill alpha-park/genpark-agent-cryptographic-audit-trail-merkle-sealer-skill This MCP server enables real-time injection fuzzing, exfiltration canary detection, least-privilege tool sandboxing, and cryptographic Merkle audit proofs for agent prompts, RAG context, and actions. It provides tamper-evident SHA-256 tree | |
| genpark-agent-deadlock-liveloss-loop-detector-skill alpha-park/genpark-agent-deadlock-liveloss-loop-detector-skill Enables monitoring and interruption of autonomous agent runs by detecting infinite tool loops, livelocks, and planning deadlocks, with safety guardrails for prompt injection, spend limits, and sandboxed command execution. | |
| Deep Research Fact-Checker alpha-park/genpark-agent-deep-research-fact-checker-skill Enables autonomous agents to cross-reference factual claims against multiple source documents, detect ungrounded or hallucinated statements, and return normalized confidence scores via the verify_claim tool. | |
| Document AST Parser & Compiler Skill alpha-park/genpark-agent-document-ast-parser-compiler-skill Enables autonomous agents to parse markdown into structured AST nodes and compile modified ASTs back into clean canonical markdown, including headings, lists, tables, and code blocks. | |
| genpark-agent-dynamic-few-shot-example-selector-skill alpha-park/genpark-agent-dynamic-few-shot-example-selector-skill Enables MCP-compliant clients and autonomous agents to pick optimal few-shot in-context exemplars by combining semantic token overlap, cosine similarity, and Maximal Marginal Relevance diversity ranking for prompt optimization. Runs determi | |
| genpark-agent-dynamic-persona-style-adapter-skill alpha-park/genpark-agent-dynamic-persona-style-adapter-skill Enables agents to adjust message formality, brevity, tone, and technical depth based on channel context, user cognitive load, and sentiment feedback. It runs as a zero-dependency, MCP-compatible server for consistent communication across in | |
| genpark-agent-execution-tree-rollback-checkpointer-skill alpha-park/genpark-agent-execution-tree-rollback-checkpointer-skill Enables autonomous agents to snapshot execution-tree and context states, evaluate branch viability, and roll back to previous checkpoints when an exploration path fails without restarting from scratch. | |
| Executive Brief & KPI Dashboard Synthesizer alpha-park/genpark-agent-executive-brief-dashboard-skill Enables autonomous agents to transform raw telemetry, system metrics, and departmental status updates into leadership-ready decision briefs, KPI watchpoint summaries, strategic signal extraction, and actionable mitigation recommendations. | |
| GenParkAgentExfiltrationCanaryLeakDetector alpha-park/genpark-agent-exfiltration-canary-leak-detector-skill Enables real-time injection fuzzing, exfiltration canary detection, least-privilege tool sandboxing, and cryptographic audit proofs for untrusted prompts, RAG context, and agent actions over MCP. | |
| genpark-agent-hallucination-fact-verifier-skill alpha-park/genpark-agent-hallucination-fact-verifier-skill Enables AI agents and MCP clients to decompose text into atomic claims and cross-verify them against reference knowledge to flag ungrounded claims, numerical contradictions, and hallucinated entities. Runs deterministically with zero depend | |
| genpark-agent-hierarchical-episodic-memory-skill alpha-park/genpark-agent-hierarchical-episodic-memory-skill Enables LLM clients to manage hierarchical working, dialogue, and episodic memories with decay-aware consolidation, hybrid lexical/vector retrieval, knowledge-graph expansion, and semantic caching over MCP. | |
| genpark-agent-human-in-the-loop-approval-gate-skill alpha-park/genpark-agent-human-in-the-loop-approval-gate-skill Enables autonomous agents and multi-agent frameworks to route actions through deterministic human approval gates before execution. Enforces risk tiers, reversible checkpoints, and token approval bounds so that risky or costly operations can | |
| genpark-agent-hybrid-memory-retriever-skill alpha-park/genpark-agent-hybrid-memory-retriever-skill Enables MCP-compatible agents and IDEs to retrieve episodic long-term memories using hybrid BM25 keyword search, semantic n-gram cosine similarity, and recency decay ranking, with deterministic zero-dependency execution. | |
| genpark-agent-json-schema-auto-repair-skill alpha-park/genpark-agent-json-schema-auto-repair-skill Enables agents and MCP clients to deterministically extract and repair JSON/schema issues such as truncated braces, markdown wrappers, and invalid types without LLM round-trips. | |
| genpark-agent-long-horizon-context-compactor-skill alpha-park/genpark-agent-long-horizon-context-compactor-skill Enables deterministic, zero-dependency long-horizon conversational memory compaction and episodic anchor extraction for AI agents, with native MCP protocol support and structured JSON telemetry output. | |
| GenParkAgentMCPToolPermissionLeastPrivilegeAuditor alpha-park/genpark-agent-mcp-tool-permission-least-privilege-auditor-skill Enables real-time injection fuzzing, exfiltration canary detection, least-privilege tool sandboxing, and cryptographic audit proofs through the Model Context Protocol. | |
| Meeting Minutes & Action Extractor MCP Server alpha-park/genpark-agent-meeting-minutes-action-extractor-skill Enables parsing meeting transcripts into structured minutes with consensus decisions, assigned action items, owners, deadlines, and follow-up agendas through the extract_meeting_minutes tool. | |
| genpark-agent-multimodal-document-table-extractor-skill alpha-park/genpark-agent-multimodal-document-table-extractor-skill Enables AI agents and MCP clients to heuristically parse multi-column document dumps, aligned spacing, and CSV/TSV input, converting them into clean GitHub Flavored Markdown tables and typed JSON. Runs fully offline on the Python standard l | |
| PRD & Technical Spec Synthesizer alpha-park/genpark-agent-prd-spec-synthesizer-skill Enables autonomous agents to turn high-level project goals into structured user stories, verifiable Gherkin acceptance criteria, edge cases, and OpenAPI-compatible API contract skeletons through MCP tools. | |
| Prompt Guard Injection Sanitizer Skill alpha-park/genpark-agent-prompt-guard-injection-sanitizer-skill Enables agents and applications to detect, inspect, and neutralize prompt injection, jailbreak attempts, and hidden instruction escapes in untrusted text before it is passed into prompts. | |
| genpark-agent-prompt-injection-jailbreak-sentinel-skill alpha-park/genpark-agent-prompt-injection-jailbreak-sentinel-skill Lets MCP-compatible assistants vet inbound prompts and untrusted text for instruction-override manipulation, separator escapes, and concealed encodings before processing them. Additional inline controls cover spending limits, repetitive-pla | |
| genpark-agent-prompt-injection-sanitizer-firewall-skill alpha-park/genpark-agent-prompt-injection-sanitizer-firewall-skill Enables autonomous AI agents and MCP-compliant clients to detect and block adversarial prompt injections, jailbreak attempts, system prompt exfiltration, and tool hijacking via a zero-dependency, sub-millisecond heuristic firewall. | |
| genpark-agent-rate-limit-token-bucket-regulator-skill alpha-park/genpark-agent-rate-limit-token-bucket-regulator-skill Enables AI agents and MCP-compatible clients to enforce multi-provider token-bucket and sliding-window rate limits, preventing TPM/RPM throttling across LLM APIs with predictive token consumption budgeting. | |
| genpark-agent-recursive-self-critique-reflection-skill alpha-park/genpark-agent-recursive-self-critique-reflection-skill Enables MCP clients to run deterministic, zero-dependency recursive self-critique and reflection loops over a payload and optional options, iteratively evaluating and revising agent output to autonomously repair logic defects, test failures | |
| GenParkAgentRedTeamIndirectInjectionFuzzer alpha-park/genpark-agent-redteam-indirect-injection-fuzzer-skill Enables real-time indirect prompt injection fuzzing, exfiltration canary detection, least-privilege tool sandboxing, and cryptographic audit proofs over MCP. | |
| genpark-agent-runtime-self-healing-diagnostics-skill alpha-park/genpark-agent-runtime-self-healing-diagnostics-skill Enables autonomous agent runtime diagnostics and self-healing by detecting configuration drift, corrupted execution state, and missing runtime resources, then classifying root causes and executing snapshot rollback. | |
| Sales Pitch & Objection Handler Skill alpha-park/genpark-agent-sales-pitch-objection-handler-skill Enables autonomous enterprise sales agents to generate customized B2B sales proposals, ROI framing, competitive battlecard talking points, and objection-handling scripts for budget, security, compliance, and migration resistance, along with | |
| genpark-agent-sandboxed-virtual-code-runner-skill alpha-park/genpark-agent-sandboxed-virtual-code-runner-skill Enables agents and MCP-compliant clients to run multi-language code in a deterministic, zero-dependency sandbox where AST security analysis blocks dangerous syscalls and isolated execution telemetry is captured. It exposes this through JSON | |
| genpark-agent-self-consistency-consensus-scorer-skill alpha-park/genpark-agent-self-consistency-consensus-scorer-skill Aggregates multiple divergent LLM reasoning trajectories through majority voting and consensus scoring to produce high-confidence deterministic outputs without additional model calls. Plug into any MCP-compliant client to resolve agent task | |
| Semantic Cache Query Matcher alpha-park/genpark-agent-semantic-cache-query-matcher-skill Enables agents to store LLM query-response pairs and retrieve them via fuzzy cosine similarity matching, reducing duplicate inference calls; it exposes tools for setting cache entries, semantic lookup, and hit-rate statistics. | |
| genpark-agent-semantic-cache-similarity-deduplicator-skill alpha-park/genpark-agent-semantic-cache-similarity-deduplicator-skill Enables agents to cache and deduplicate semantic queries using exact SHA-256 matching and cosine similarity thresholds with LRU eviction, reducing redundant LLM calls and improving throughput. | |
| genpark-agent-session-compaction-tokenizer-skill alpha-park/genpark-agent-session-compaction-tokenizer-skill Enables autonomous agents to tokenize session context windows and compact dialogue history deterministically, exposing this capability to MCP-compatible clients like Claude Desktop and Cursor. Helps manage context length and reduce token us |
26,851–26,900 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.