Reglint is the runtime compliance layer for AI agents. Security tools catch jailbreaks and prompt injections. Reglint catches what the law actually requires — discrimination, fraud framing, unauthorized advice, PII exposure, fiduciary violations, and behavioral crimes — and maps each to the specific statute violated. ━━━ TWO LAYERS, ONE UMBRELLA ━━━ → Code Scan (Layer 1): Pre-deployment GitHub Action that flags AI compliance issues before code ships. → Agent Monitor (Layer 2): Runtime API that scans every AI agent output in 500ms (Fast) or 3-5s (Full) and returns BLOCK / REDACT / ALERT / PASS with the specific regulation cited. ━━━ COVERAGE ━━━ 27+ violation patterns across 5 industries: - Finance — GLBA, BSA, OFAC, FCRA, Reg E, ECOA - Healthcare — HIPAA, HITECH, 42 CFR Part 2 - HR / Hiring — ADEA, Title VII, NYC LL144, Colorado SB205 - Education — FERPA, COPPA, IDEA - General — GDPR, CCPA, EU AI Act, GINA, TCPA ━━━ INTEGRATION ━━━ 5-minute REST API. Works with any AI agent — LangChain, OpenAI SDK, Anthropic SDK, n8n, custom stacks. ━━━ FOR ━━━ SMB and federal teams deploying AI agents who need defensible compliance — not just dashboards. Production live on AWS-native infrastructure. SAM.gov registered (UEI QRWWCLHF4JB1). Virginia LLC. Active U.S. Public Trust Clearance on the founding team. Arlington, VA — reglint