Free, keyless French company checks (Sirene, BODACC, VIES, asset freezes) and DPE rental compliance.
Check before paying (x402), email/domain checks, vendor security, company research.
Scan packages and lockfiles (npm, PyPI, Go, Maven, Cargo, NuGet) for vulnerabilities and malware.
Companies House, Gazette notices, sanctions screening, LEIs and UK legislation.
Open-source local stdio MCP runtime for agents. Applies policy and scoped approvals to tool calls, verifies outcomes, and reconciles interrupted actions instead of blindly replaying them. Alpha.
Read-only access to your Auditon vulnerability scans: findings, endpoints, reports and changes.
UAE ICV compliance: governed scores, evidence gaps and actions with deterministic numbers.
Read and update your 21RISK data: sites, audits, compliance, actions, claims and insurance values.
Public holidays, business days, VAT, withholding tax and tax-ID checks for 54 African countries.
Security basics of a website: HTTPS certificate, redirection, headers, mixed content, DNS filtering.
What changed in your stack since the model's cutoff, error fixes, and security-scanned skills/MCPs.
Indian FEMA compliance: RBI purpose codes, EDPMS / IDPMS, Form 145 / 146 and export deadlines.
Read-only search of a curated directory of ISO 27001, SOC 2 and compliance services.
DMARC/SPF/DKIM/MX checks, email score, 100-domain bulk score, webhook watch alerts (x402 USDC).
Optimal Social Security claiming, in minutes. Benefit-by-age math from a deterministic engine.
Free website score for speed, SEO, security and AI search readiness, with the fixes.
About Security MCP servers
Security servers let agents take part in defensive work: scanning dependencies and containers, querying vulnerability databases and threat intelligence feeds, reading audit logs, managing secrets in vaults, checking identity and access configuration, and producing compliance evidence. Several wrap established tools so an assistant can run a scan and explain the findings in plain language.
These servers are powerful on both sides, so run them with the same controls you apply to the underlying tools: dedicated credentials, read-only where possible, and no production write access without a human in the loop. Prefer servers from the tool's own vendor or an established security team, and review what each tool sends to third-party services.