Cequence Platform 9.0 adds AI assistant for API security

Cequence Security has launched Cequence Platform 9.0, adding an AI-native interface for managing API security tasks.

The release includes a built-in AI Assistant and an open Model Context Protocol server that lets customers use either Cequence's assistant or another AI agent to query and act on security data.

Users can ask plain-language questions and complete tasks such as classifying APIs, identifying risks, drafting rules and creating reports without navigating the platform's user interface. Organisations can also connect their own agents, SOAR tools or automation workflows through the MCP architecture without custom integration.

Every proposed write action still requires human approval before any change is made. The assistant also shows its reasoning and underlying tool calls, and states when it lacks the means to complete a task.

Compliance focus

Compliance is a central part of the new release. Platform 9.0 includes more than 250 pre-built risk rules mapped to 25 regulatory and compliance frameworks, including OWASP API Security Top 10, PCI DSS, GDPR, HIPAA, SOC 2, ISO 27001, NIST CSF, DORA and NIS2.

The system can generate audit-ready reports from live data, mapping findings to specific controls, scoring risk by control area and providing remediation guidance for gaps. It also includes an observe mode for testing proposed rules without raising formal issues, and a test panel for validating rules against sample request and response data before activation.

Cequence is targeting organisations with growing API estates and increasing pressure to demonstrate compliance across multiple jurisdictions. The launch reflects broader demand for tools that fit into AI-driven workflows while keeping security teams in control of changes.

Scale changes

Alongside the AI features, Cequence has rebuilt the engine that discovers, catalogues and scores risk across an organisation's API estate. The updated engine supports 50 times more API endpoints than the previous version while keeping page load times below five seconds across every view.

The redesign also reduces CPU use, which could lower infrastructure costs for customers running the platform in their own environments.

Ameya Talwalkar outlined the architectural shift behind the release.

"Most vendors looked at the agentic era and added a chatbot. We looked at it and rebuilt the architecture. Cequence Platform 9.0 exposes the entire Cequence platform through an open MCP architecture so any agent can operate it directly, whether through our built-in AI Assistant, or a customer's own agent. That is what AI-native actually means: the UI becomes optional. We are building for the way the agentic enterprise already works, while making sure a human approves every change along the way," said Ameya Talwalkar, Chief Executive Officer and Co-Founder, Cequence.

Cequence positions its approach as different from products that layer a chat interface on top of existing security tools. Instead, it presents the platform as one where conversational access and agent-based automation are part of the core design.

Shreyans Mehta said the aim was to make the system useful beyond specialist security users.

"Most security chatbots are only as useful as the person asking the questions, which means they fall flat in the hands of anyone who is not already an expert. We built the Platform 9.0 agent differently. It runs a full agentic loop, planning which tools answer the question, calling them, and synthesising ranked, evidence-backed recommendations while showing you exactly how it got there. When it does not have the tool to do something, it tells you instead of guessing. That governance-first design is not an afterthought. It is the same conviction behind the Cequence AI Gateway, and it is what makes this safe to put in front of any practitioner on Day 1," said Mehta.

Cequence focuses on application, API and AI-related security. Its platform protects more than 10 billion daily API interactions and 4 billion user accounts.