SmartBear releases BearQ agentic QA system
SmartBear today announced the release of BearQ, an agentic QA system that can discover how applications work to ensure their integrity.
According to the company, BearQ, a part of SmartBear’s Application Integrity Core product suite, can autonomously explore how applications work and adapt to any changes in code, testing only what matters. The ability to adapt makes this solution an evolution fron brittle automation scripts or predefined test cases.
“The AI-driven gap between the speed of software development and the ability to test resulting applications is getting wider, creating a huge potential for application failures that hurt companies, consumers and industries,” said Dan Faulkner, SmartBear CEO, in the announcement. “BearQ will be a critical autonomous exploration and testing solution for the industry to close these gaps and deliver enhanced application integrity. SmartBear’s long history of innovation in software quality and testing makes SmartBear positioned to address this disruption to the SDLC.”
A recent SmartBear study, Closing the AI Software Quality Gap, found that 90% of respondents are using AI coding tools, but 70% say software quality is suffering. Further, more than three-quarters of respondents are worried that testing won’t be able to keep up with faster AI code development, nine of 10 survey respondents have adopted AI coding tools. Seven of 10 are concerned that quality is already suffering, 60% have experienced quality issues in the past year, and 68% are worried that testing will be a bottleneck for software delivery.
BearQ also includes guardrails and contextual controls that allow developers to define what agents must validate, and how exploration and testing align with business goals.
To learn more about BearQ and apply for GA Early Access, click here.
Chainguard announces open source Agent Skills
Chainguard today announced Chainguard Agent Skills, a catalog of hardened AI agent skills that is continuously maintained. Agent Skills, which automatically ingest skills from open source registries, enables developers to expand the use cases of their agents without extending their attack surface. Those skills are measured against a security and quality ruleset, hardens them using Chainguard reconciliation agents, and publishes them with a complete audit trail. That vetting is important, because without permission controls or oversight, agent skills have become the latest target of supply chain attacks.
AI agent skills are small, modular instruction sets that extend what an AI agent can do. Developers install skills to add capabilities such as browser automation, PDF processing, database access, and code-generation workflows. agent skills have become the latest target of widespread supply chain attacks.
“Container images showed us how quickly software artifacts can become supply chain risks once they’re adopted and trusted at scale. AI agent skills are emerging along an even faster trajectory,” said Dan Lorenc, CEO and Co-founder, Chainguard. “As AI agents become embedded in the software development lifecycle, the skills that shape their behavior become part of the supply chain itself. With Agent Skills, Chainguard is bringing continuous hardening and verifiable integrity to that layer, so organizations can build with AI on a secure foundation.”
Chainguard Agent Skills is available in beta. To be among the first to try Chainguard Agent Skills, visit https://www.chainguard.dev/agent-skills.