Tricentis Launches End-to-End Enterprise Agentic Quality Engineering Platform
Software QA provider Tricentis has released a new-powered agent quality engineering platform Tricentis AI workspace, which works by deploying AI agents to manage risk while enabling rapid innovation.
AI has changed the face of software development and implementation in terms of the speed and scope of application changes, and discovered errors are just one application in a connected ecosystem that can introduce risk, increase downtime and damage a business.
In the announcement, Tricentis wrote: “Tricentis AI Workspace serves as a command center with a shared context, integrated workflow and traditional agent-to-agent collaboration to act as a system of record and ‘control tower’ for agent quality engineering, connecting AI agents to all testing, automation, operational and quality intelligence, while embedding governance, authorization and auditing directly.”
Kevin THompson, CEO of Tricentis, noted that “AI is changing in its ability to create code at an unprecedented speed, but the conflict caused by lack of confidence in the quality of the product causes real pain for CIOs. Although businesses want speed, they also cannot bring risks by using insecure or inferior code generated by AI. “The problem of Tricentis today. We offer the first end-to-end software quality platform that redefines how enterprise software can be tested, managed, and released to deliver high-quality code at the speed of AI while safely accelerating time-to-value.”
Within the Tricentis AI Workspace there are several AIs that work together with defined responsibilities throughout the software development life cycle (SDLC):
- Tricentis Agetic Quality Intelligence: Continuously interprets change, risk, and quality signals throughout the SDLC to determine readiness for release, automatically directs testing and escalates to humans only when judgment is needed.
- Tricentis Agetic Test Automation (updated): Building on the first launch of Agentic Test Automationthis next generation increases productivity. New features include support for SAP GUI and web applications, deep integration with Tricentis Tosca automation engines, and intelligent reuse of test modules to reduce repetition, maintenance, and risk.
- Performance evaluation of Tricentis Agetic: It delivers business-ready, AI-driven operational assurance by embedding autonomous agents throughout analysis, design, and execution – accelerating insights by up to 90-95%, removing manual expert bottlenecks, and enabling faster, more confident AI release decisions from API to end-to-end applications.
- Tricentis Agetic Test Creation: Deeply integrated into Tricentis qTest, Agentic Test Creation sits alongside test developers, helping them with in-content test validation. It enables natural language test creation, allowing teams to create reusable test cases quickly and consistently while reducing duplication and reliance on specialized knowledge.
Mend.io introduces System Prompt Hardening
Mend.io introduced System Prompt Hardening within Mend AI, designed to improve weaknesses in AI system prompts. System information often contains hidden instructions, which has become a security concern that traditional AppSec tools cannot address. System Prompt Hardening provides visibility into these commands, identifies vulnerabilities and automatically hardens quickly to mitigate risk before applications reach production, according to the company.
According to Gartner, 32% of organizations have reported attacks on AI applications that have rapidly implemented an application within the past year.
“System alerts are the blueprint for AI applications, but security standards are not keeping up with their growing importance,” said Rami Sass, GM of Mend AI. “While security and development teams have established frameworks such as CWE and CWSS for software vulnerability assessment, we are now introducing System Prompt Hardening and AIWE (in closed beta testing) as the first official methods for testing and improving these prompts.”
Together, Mend.io’s System Prompt Hardening and AIWE represent a paradigm shift in how organizations manage AI risk. Key skills include:
Automatic Detection and Labeling of Content: A possible context for security and development teams with continuous identification of hidden system information that is automatically classified by its function and possible attack vectors;
Systematic Severity Score: A proprietary one-to-one severity scale that accurately measures the risk of specific immediate risks and prioritizes remediation based on measurable impact; again
Remediation – Operating System Rapid Resilience: In addition to easy detection, the platform automatically suggests improvements to inform logic to mitigate threats such as rapid injection before applications reach production.
System Prompt Hardening is available for Mend AI Core and Mend AI Premium, extending the integrated platform to protect AI-generated code and embedded AI components, reduce risk with AI-powered remediation, automate compliance, and provide enterprise-scale visibility into application risk across the development lifecycle.
