How to report a security issue.

This policy covers security vulnerabilities in the Prism service at prismlens.net, including the web application, API, and any mobile clients we release. It does not cover third-party services (Supabase, Vercel, Anthropic, OpenAI, Stripe, Resend) — report issues in those directly to the vendor.

·Authentication and authorisation flaws (access control bypasses, session handling, privilege escalation).

·Data exposure or leakage (other users’ entries, embeddings, analysis, or personal data visible without authorisation).

·Injection flaws (SQL, template, command, prompt injection that leaks data across users).

·Remote code execution.

·Server-side request forgery (SSRF) or other infrastructure risks.

·Logic flaws that allow free access to paid features, bypass of rate limits, or financial loss.

·Supply-chain risks specific to Prism’s deployment.

·Third-party issues in Supabase, Vercel, Anthropic, OpenAI, Stripe, Resend — report to the vendor.

·Denial of service, volumetric attacks, or physical attacks.

·Social-engineering attacks against Prism staff.

·Self-XSS requiring the user to paste code into their own browser.

·Missing security headers with no demonstrable impact.

·Clickjacking on pages with no sensitive action.

·Issues requiring an already-compromised device or root/jailbreak.

·CSRF on endpoints with no security impact.

·Rate limiting / brute force without credential-valid impact.

·Reports generated entirely by automated scanners with no manual verification.

Email security@prismlens.net with:

·A clear description of the issue.

·Steps to reproduce, or a proof-of-concept.

·Your assessment of impact (who is affected, how).

·Your contact details (optional) if you want attribution or a reply.

·Acknowledge receipt within 3 business days.

·Triage and respond with our initial assessment within 7 business days.

·Keep you informed through remediation.

·Not take legal action against good-faith research conducted within this policy, provided you avoid accessing or exfiltrating other users’ data beyond the minimum needed to demonstrate the issue, do not degrade service availability, give us a reasonable window (default 90 days) before public disclosure, and comply with applicable law.

·Don’t test against real users’ data. Create a test account.

·Don’t use social engineering against Prism’s founders, staff, or users.

·Don’t exfiltrate data beyond what’s needed to prove the issue.

·Don’t publicly disclose before we’ve had a reasonable window to respond and remediate.

·Report in English or Arabic.

Prism does not currently operate a cash bounty programme. For significant issues we’ll acknowledge publicly (with your permission), credit you in a changelog entry, and issue Prism swag where available. Our priority is remediation speed, not a bounty budget — we will revisit as we grow.

To the fullest extent we can offer, your good-faith research under this policy:

·Is authorised by Prism and does not violate the Prism Terms of Service.

·Will not be grounds for civil or criminal action by Prism under applicable computer-misuse laws.

·Will not result in a complaint to your employer, regulator, or law enforcement.

This safe harbour is our commitment and does not guarantee that third parties (payment processors, infrastructure providers) will take the same view. If in doubt, ask us before you test.

·Security reports: security@prismlens.net

·General support: support@prismlens.net

·Privacy / DPO: privacy@prismlens.net · dpo@prismlens.net