Osiris Labs provides an agentic email security platform that automatically detects, removes, and documents malicious emails across Microsoft 365 and Google Workspace. The platform uses multiple AI agents working in parallel to analyze every inbound message and remediate confirmed threats without human intervention.
Funding
Funding not disclosed
Founders
Product
Problem
Traditional email security tools generate alerts for threats but leave the actual remediation work to security analysts, who must manually investigate and remove emails from every affected inbox. This creates a significant time gap between detection and resolution, during which malicious emails remain accessible to users and can cause further damage. Additionally, standard security tools provide only risk scores rather than complete forensic evidence, leaving security teams without the full context needed for breach investigations.
Solution
Osiris Labs delivers an agentic email security platform that autonomously investigates and neutralizes email threats across Microsoft 365 and Google Workspace environments. The platform deploys three specialized AI agents: HORUS analyzes every inbound email across multiple detection signals simultaneously, TALON automatically quarantines and sweeps confirmed malicious messages from all inboxes within seconds, and ATLAS compiles complete forensic records for every verdict. The system classifies message intent at the semantic level, unpacks attachments recursively, decodes QR codes, and traces URL chains through multiple redirects. Every verdict includes a full evidence chain with detection signals, URL resolution paths, and remediation actions, providing security teams with case files rather than simple risk scores.
Target Audience
Primary customers are security teams and IT administrators at enterprises using Microsoft 365 or Google Workspace who need automated email threat detection, remediation, and comprehensive forensic reporting.
Features
- Multi-Vector Detection Engine that runs all detection signals simultaneously on every email, eliminating sequential check gaps
- NLU Intent Analysis that classifies social engineering patterns, financial pressure language, urgency cues, and impersonation signals at the semantic level
- Autonomous remediation that sweeps confirmed threats from every inbox in the organization within seconds without analyst intervention
- Complete forensic records for every verdict, including URL chain traces, sender analysis, and action logs
- Outbound email classification that scans and categorizes all outbound messages by content type, data category, recipient context, and sharing patterns
- Automated security reporting that compiles organizational threat data, authentication health across SPF, DKIM, and DMARC, and trend comparisons without manual data pulls