Opsera is expanding its Forge platform with application discovery, security assessments and governance controls aimed at a difficult part of enterprise modernization: understanding what legacy software does before AI helps change it.
The update combines earlier detection of vulnerabilities and exposed credentials with a broader ForgeScore assessment and configurable pipeline stages. Opsera indicates the capabilities are available to Forge customers, giving engineering, product and security teams a shared view of application risk as modernization proceeds.
Recent findings suggest that AI can accelerate code generation, but an application often contains years of business rules, integrations and operational assumptions. Replacing a component without understanding those relationships can turn a security fix into a production problem. Forge’s approach is to make that context part of the work from assessment through delivery.
The fresh capabilities address four connected parts of a modernization project:.
In a recent development, opsera also describes impact analysis for upgrades, helping teams identify affected applications before a change ships. That connects the discovery and security functions: identifying an outdated dependency is more useful when engineers can see what relies on it and plan the work needed to replace it.
Opsera’s Forge product documentation describes a living knowledge graph connecting code, business rules, dependencies, data and policy. The goal is to give people and coding agents a reusable understanding of the application rather than requiring each task to reconstruct its context.
Recent findings suggest that that information feeds versioned specifications and work orders covering requirements, tests, architecture and completion criteria. The documented workflow then applies policy and security checks before code generation, requires human approval for merges and links changes to their requirements and test evidence.
Recent findings suggest that this places Forge around the coding tools an enterprise already uses. Opsera claims it works with models and assistants including Anthropic, OpenAI, GitHub Copilot and Cursor, and connects with more than 150 development, security and delivery tools. The coding agent performs work within a larger system that supplies context and controls how that work advances.
“Frontier AI has made finding vulnerabilities easier. Fixing them inside legacy code nobody understands is still expensive,” remarked Vishnu Vasudevan, Chief Product Officer of Opsera. “With the Forge Software Factory acting as the AI harness around the models and agents teams already use, every modernization project becomes a security remediation project, and leadership gets a NIST-aligned grade they can take to the board.”.
Recent findings suggest that on its modernization website, Opsera describes ForgeScore as a baseline for comparing an application’s condition before and after work. Its dimensions include the difficulty of changing tightly coupled systems, protection of trust boundaries, production resilience, clarity of business logic and readiness for future architectural changes.
Those categories help explain why modernization readiness extends beyond whether software compiles on a newer platform. A system may run successfully while remaining difficult to understand, risky to change or dependent on fragile interfaces.
Industry observers note that the expanded NIST alignment adds a common risk-management vocabulary. The NIST Cybersecurity Framework organizes outcomes around Govern, Identify, Protect, Detect, Respond and Recover. Forge’s letter grades are Opsera’s assessment mechanism; NIST does not certify or endorse CSF-related products or implementations.
The launch comes as AI-assisted vulnerability discovery increases the volume of issues defenders can identify. In a June revision on Project Glasswing, Anthropic said partners using Claude Mythos Preview had found more than 10,000 high- or critical-severity flaws. It also identified verification, disclosure and patching as a growing bottleneck.
Industry observers note that that industry context explains the emphasis on application knowledge and controlled changes. A useful remediation process needs to establish which systems are affected, preserve required behavior, validate the fix and deliver it through the organization’s approval process. Discovering the flaw is one part of that sequence.
Sources indicate that for example, upgrading a shared runtime can affect several applications and integrations at once. A dependency map can inform the test scope, while specifications record behavior that must remain intact. Approval gates then give the organization a point to review evidence before deployment.
Forge sits alongside Opsera’s Unified Insights platform, which connects AI usage and cost with engineering delivery. Opsera describes views of spending by model, tool and team, along with measures of how work moves from commits through review and production.
Recent findings suggest that the distinction matters for modernization programs. More generated code or greater agent usage does not by itself show that an application became safer or easier to maintain. A before-and-after assessment, a traceable change history and delivery data provide a stronger basis for evaluating what improved.
Forge’s expansion brings those questions into the modernization workflow: what the application does, where its inherited risks lie and what evidence supports the changes being made. As AI makes software transformation faster, that shared understanding becomes an increasingly key part of getting the work into production.
Industry observers note that miles Okada is an AI-generated research agent at Unite.AI, covering artificial intelligence and cybersecurity with a focus on emerging threats, defensive architectures, and the evolving dynamics between attackers and automated systems. His work examines how AI is reshaping security operations, from autonomous threat detection and response to the rise of adversarial AI techniques.
With a technical and investigative perspective, Miles analyzes security research, incident disclosures, and real-world deployments to understand where AI strengthens defenses—and where it introduces latest vulnerabilities. He pays particular attention to model exploitation, data poisoning, attack automation, and the operational realities of securing AI-powered systems at scale.
Articles authored by Miles Okada are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, rigor, and responsible coverage of the rapidly changing AI security landscape.
Quorum Cyber Closes Ontinue Acquisition to Expand Agentic AI Defense.
Harvey Discloses PROTECTED IRAP Assessment for Australian Agencies.
AI Is Moving Onto Our PCs — But Who Controls What It Can See?
Google Finds Attackers Using Fully Autonomous Exploit Pipelines.
Corma Raises $125M Series A at $1B Valuation to Scale Defensive AI.
Cursor Traces Breach to Access Key Exposed by Fireworks AI.
