Security and trust
CivoraIQ is legislative monitoring software used by professional and trade associations. This page describes how the service is built and operated. It is written to be checked, not to impress — if something here is unclear, ask us at info@civoraiq.com.
On this page:Accounts and authenticationData and hostingAI featuresApplication securityService providersReporting a security issueWhat we don’t claim
Official sources only
Bill data comes from state legislatures and Congress through Open States and LegiScan. Every bill links to its official page so anything can be verified at source.
Humans approve what leaves
AI-drafted briefs, positions and report summaries are labelled as such and are reviewed by your team. Nothing AI-generated is emailed to anyone without a person approving it first.
Encrypted in transit
All traffic uses HTTPS with HSTS; plain HTTP requests are redirected. Emails are sent through an authenticated provider with SPF and DKIM on our domain.
Role-based access
Each account belongs to one organization and sees only that organization's reviews, positions, outreach and reports. Permissions are enforced on the server for every action.
Accounts and authentication
- Accounts are created by your organization's administrator; there is no open sign-up.
- Passwords are stored only as salted scrypt hashes and are never logged or emailed.
- Sessions use a signed, HTTP-only cookie. Changing or resetting a password signs out every other device.
- Repeated failed sign-ins are rate-limited per account and per network address.
- Password resets and email verification use single-use, time-limited links.
Data and hosting
- The application and database are hosted on Railway in the United States (San Francisco region). The database is on a private network and is not exposed to the public internet.
- Each organization's content is isolated by organization in the database and in every query.
- Legislative data (bills, texts, actions) is public information; your organization's profile, notes, positions and emails are not shared with other customers.
- We keep account and organization content while the account is active and delete or anonymise it on request when an organization leaves — see the privacy policy.
AI features
- AI is used for impact briefs, draft positions, Advisor answers and optional report summaries. Requests go to model providers through OpenRouter and include only what the feature needs: the bill text, your organization's profile and your question.
- Output is marked as AI-generated in the product and in emails, and is meant to be checked against the official bill text that is always one click away.
- AI summaries in scheduled reports can be switched off per report; when on, every edition waits for a reviewer.
- We do not train models on your data.
Application security
- Security headers: HSTS, frame denial, strict referrer policy, no MIME sniffing; the app cannot be embedded in other sites.
- Every write goes through server-side authorization; identifiers in links and forms are re-checked against the signed-in user's organization.
- Uploads (product-board images) are restricted by type and size and stored privately.
- Dependencies are kept current; the codebase is reviewed for security issues as part of normal development.
Service providers
We rely on a small number of providers, each for one purpose. The privacy policy lists what each receives.
- Railway
- Hosting and database (United States)
- Resend
- Email delivery
- OpenRouter
- AI model access
- Langfuse
- AI quality monitoring
- Cloudflare, Hostinger
- Domain, DNS and business email
Reporting a security issue
Email info@civoraiq.com with the subject “Security report”. We acknowledge reports promptly, keep you informed while we investigate, and ask that you give us reasonable time to fix an issue before publishing it. Please do not access other customers' data while testing.
What we don't claim
CivoraIQ is an early-stage product. We do not currently hold SOC 2, ISO 27001 or similar certifications, and we say so rather than imply otherwise. If your organization needs a security questionnaire completed, write to us and we will answer it directly.