Repello AI

Repello AI is an enterprise AI security platform from Repello Inc. It helps organizations discover AI systems, test them against attack paths and enforce runtime protections across production systems and employee workstations. Its products are Inventory, ARTEMIS, ARGUS and Workstation Lens, plus open-source tools such as Agent Wiz and SkillCheck.

What It Is

Repello positions itself as security for agentic workflows, MCP connections and chatbots. The platform follows a three-phase framework: inventorize AI assets, red-team them, then turn findings into adaptive runtime guardrails.

How the Platform Is Organized

• Inventory builds a unified view of agents, models, datasets, tools and connections, including an AI Bill of Materials.
• ARTEMIS performs autonomous red teaming. It profiles a target, builds a threat model, runs multi-stage attacks and shows attack paths. Browser Mode tests AI apps through the UI, and API testing is also supported.
• ARGUS applies runtime controls that monitor AI activity and block malicious inputs, unsafe actions and policy violations.
• Workstation Lens shows what coding agents such as Claude Code, Cursor, Codex and Copilot access, which commands they run and where they send data.

Red Teaming Workflow and Outputs

Users add an AI asset by connecting endpoints, choose attack goals from a library of evolving adversarial attacks, and launch simulations. Reports include prioritized vulnerabilities, risk scores, evidence and remediation playbooks. Findings map to frameworks including OWASP, NIST, MITRE, ISO 42001 and the EU AI Act. Repello says testing covers text, image and audio across 100+ languages and can integrate with CI/CD pipelines.

Vendor Claims

Repello states it uses 15M+ evolving attack patterns and covers 270+ AI vulnerability types. It also states that ARTEMIS gives 15x more coverage than manual testing and covers all 10 OWASP Top 10 LLM categories. The site says it is SOC and ISO 27001 certified.