Skip to content

Security: UnboundedTechnologies/UnboundedTechnologies-Landing

Security

SECURITY.md

Security Policy

Reporting a vulnerability

If you discover a security vulnerability in this site or its supporting infrastructure, please report it privately so we can fix it before it's disclosed publicly.

Contact: contact@unboundedtechnologies.com

Please include:

  • A clear description of the issue
  • Steps to reproduce (URLs, payloads, request/response samples)
  • Affected commit SHA or production URL if known
  • Your assessment of severity and potential impact
  • Whether you've already shared the issue with anyone else

We aim to acknowledge reports within 2 business days and to ship a fix or mitigation within 30 days for medium / high severity issues. We will credit reporters in the release notes if you'd like (let us know in the report).

Please do not open public GitHub issues for security disclosures.

Scope

In scope:

  • Production site at https://unboundedtechnologies.com
  • Source code in this repository
  • The contact funnel (/api/contact) and its third-party integrations as configured (Notion, Resend, Upstash, Calendly)

Out of scope:

  • Findings from automated scanners that report informational issues without proof of exploitability (rate-limit guesses, CSP "could be tighter", etc.)
  • Social-engineering attempts against owner accounts
  • Vulnerabilities in third-party services themselves (please report directly to Vercel / Cloudflare / Notion / Resend / Calendly / Upstash)
  • Pre-deployment branches, preview URLs, or unmerged PRs

Hardening already in place

  • Content-Security-Policy with a tightly scoped URL allowlist. script-src uses 'unsafe-inline' plus a small set of trusted origins (Calendly, Plausible, Vercel analytics). Per-request nonces were dropped after Next 16 emitted "nonce":"$undefined" for notFound() flight payloads, which broke the 404 page; per CSP3 modern browsers ignore 'unsafe-inline' when a nonce is present, so the "fallback" workaround would not actually relax modern enforcement. XSS surface is controlled by object-src 'none', base-uri 'self', form-action 'self', frame-ancestors 'none', and a no-DOM-echo funnel
  • HSTS with max-age=63072000; includeSubDomains; preload
  • X-Frame-Options: DENY plus CSP frame-ancestors 'none'
  • X-Content-Type-Options: nosniff
  • Referrer-Policy: strict-origin-when-cross-origin
  • Permissions-Policy denying camera / microphone / geolocation / payment / usb / sensors
  • Cross-Origin-Opener-Policy: same-origin
  • Hidden hp_field honeypot input on the contact form (visually hidden + tab-skipped). Humans never fill it; naive bots blindly fill every input. Honeypot rejections return a 200 success-shaped response so the bot never learns the rejection mechanism
  • Upstash sliding-window rate limit (3 form submissions per IP per hour) on /api/contact
  • Server-side validation of all funnel inputs via Zod (leadSchema)
  • All third-party clients (Notion, Resend, Upstash) soft-fail if their env var is absent so a missing secret never leaks an internal error to the user; Promise.allSettled around CRM + email writes means a single integration outage cannot 500 the route

Note on Turnstile: Cloudflare Turnstile (invisible mode) was previously deployed on the contact form. It was removed in May 2026 because its always-mounted background risk-analysis ran on the main thread and produced multi-second freezes during typing/scroll on iPhone, making the form effectively unusable. If a smarter challenge is reintroduced later, it will only mount after a first failed honeypot attempt to avoid that regression.

Responsible disclosure

We follow the principle of coordinated disclosure. We will not pursue legal action against researchers who:

  • Make a good-faith effort to avoid privacy violations, destruction of data, and interruption of service
  • Limit testing to accounts they own or have explicit permission to test
  • Give us reasonable time to fix the issue before public disclosure

Thank you for helping keep our customers safe.

There aren't any published security advisories