Security MCP servers

Free website privacy scanner for pre-consent cookies, trackers, consent, policy, and HTTPS/TLS.

Local-first domain security and privacy checker: live SPF, DMARC, DKIM, DNSSEC, TLS, CAA and security header checks, scored with remediation advice.

CIRCL Vulnerability-Lookup — aggregated vulnerability records

Read-only codebase audits for coding agents: secrets, RLS, deps, review verdicts. No writes.

Turn other AIops tools' audit trails into tamper-evident, hash-chained evidence bundles mapped to HIPAA, PCI-DSS, SOC 2 and GDPR. Offline and deterministic.

Pay-per-call x402 tools (utilities, DNS/WHOIS, blockchain RPC, market and DEX data, prediction markets, security audits) plus an OpenAI-compatible LLM gateway.

Register and verify domains, queue security scans (headers, TLS, DNS, ports, CVEs and AI-specific checks) and read findings.

Signed audit chain for AI agent decisions: events are signed, Merkle-chained per project, anchored to public transparency logs and verifiable offline.

Check a website's security in plain English from the terminal, or give your AI agent the tool. Free, no key.

Deterministic security scan of public git repositories: vulnerable dependencies via OSV, hard-coded secrets and risky config lint, as file:line findings.

MCP server for FinishKit. Production readiness scanner for AI-built apps. Enables AI agents in Claude, Cursor, Windsurf, and VS Code to check if code is ready to ship.

Scan Google News and PR wires for a company's funding rounds, executive moves, launches and acquisitions as deduplicated, dated events in Clay-ready JSON.

Permission tiers, approval gates and secret-redacting audit logging enforced inside MCP servers at the tool boundary, with a demo order-desk server.

Resolve a company domain to its GitHub organization with repository, language and activity signals.

GreyNoise Community MCP — internet scanner classification (free tier with key)

Claim-level hallucination detection with sources, agent trajectory verification, risk gating for computer-use actions and prompt-injection defense.

MCP server interface for AI agent and MCP security auditing — config analysis, trust audits, prompt injection testing, tool probing, and data flow tracing

MalwareBazaar MCP — abuse.ch malware sample database (free, key required)

Accountability for AI agent fleets: verify a signed, hash-chained ledger, check budgets, kill an agent, and issue passports and mandates, all through the local CLI.

market-spread MCP — cross-venue prediction-market landscape scanner.

68k+ security-audited MCP skills + free agent trust layer: signed ATC cards, tool gating, audits.

MDN HTTP Observatory — grade any website's HTTP security headers and get the specific fixes, from Mozilla's public scanner.

MCP server for negbit — discover and read curated knowledge bundles (verified provenance, security-reviewed) agents can buy with their own x402 wallet. Read-only: never holds a wallet or moves money.

Free passive security scanning for AI agents. Check any domain for email spoofing (DMARC/SPF/DKIM), TLS, security headers, exposed files, and subdomain takeover risk — using public information only, no intrusive probing.

NVD MCP — wraps the NIST National Vulnerability Database API.

Live geolocated world-event intelligence and a daily world brief: event signals, hotspots, Sentinel satellite imagery and an OSINT/GEOINT analyst.

Privacy-first local MCP server: redact secrets, decode JWTs, estimate LLM costs, split long documents and check AI output for hallucinations — all on your own machine, with no network calls.

Pulsedive MCP — threat-intelligence IOC enrichment (pulsedive.com)

Self-hosted default-deny action registry, append-only spend-intent ledger and human sign-off audit trail for gating irreversible agent actions.

HTTP security testing toolkit (send, intruder, race, chain, out-of-band) with low-level HTTP/1.1 and HTTP/2 control: raw framing and connection pinning.

SEO, AEO, GEO, local SEO and CRO auditor with framework-aware code fixes, sitemap crawling and HTTP security header inspection.

Fail-closed policy proxy for MCP tool calls that denies, warns on or holds each call for human approval before forwarding, with client-side redaction.

U.S. exclusion, debarment and license screening (OIG LEIE, SAM.gov, state boards) with an Ed25519-signed receipt of the exact list snapshots checked.

Exactly-once payments for agents: prevents double settlement across processes and retries, confirms charges with the provider and reconciles out-of-band spend.

SEC 13F — institutional ownership, asked security-first.

SEC Form N-PORT — fund and ETF holdings, asked security-first.

MCP server that lets an agent write secrets into .env without ever seeing the value — user pastes into a native Windows masked dialog; PowerShell writes the file.

Security-focused SendGrid email sending with recipient/domain allowlists, send caps, review mode, an audit log and no silent BCC.

Analyze AWS IAM policies and CloudFormation templates with Shieldly to flag privilege-escalation paths, wildcards and over-permissive access.

Read-only access to self-hosted SonarQube Community Build: issues, security hotspots, rules and code snippets so an agent can fix findings locally.

Core Web Vitals metrics by CMS, CDN and framework, based on Chrome field data and a multi-site crawl.

Document forensics via the Tamperlens API: detect post-creation edits, failed redactions and embedded prompt injection in PDFs, Office files and images.

Tennessee DMV MCP — Department of Safety and Homeland Security: the 186 driver service

Check the reputation of AI microservices and public APIs before calling them and submit ratings afterward, via an open trust registry.

UK Health Security Agency (UKHSA) data dashboard MCP — keyless.

Trust scores and guardrails for AI-generated code: static security rules, Docker-verified build/boot checks and receipts that catch agents lying about tests.

URLhaus MCP — wraps abuse.ch URLhaus malware URL database (free, no auth)

Fail-closed trust gate for agents: fact-checking, prompt-injection detection, content moderation, PII and secret detection, and a pre-action guardrail.