Stream Security launches StreamForce AI agent platform

Stream Security said StreamForce is now generally available, giving security teams AI agents that run on a live model of their production environment instead of raw logs or hand-built context. The company says the platform is designed to speed up security workflows while reducing the infrastructure and integration work usually required to deploy AI in operations.

Why it matters: - Security teams can now build AI workflows on a live environment model rather than reconstructing context from logs or static data. - Stream Security is positioning StreamForce as a way to reduce the setup burden that has slowed AI adoption in security operations. - The platform is meant to help teams act on current reality in cloud, on-prem, SaaS, identity, ephemeral assets, and AI agents in production.

What happened: - Stream Security announced general availability of StreamForce, its AI agent orchestration platform, on July 9, 2026. - StreamForce is built natively on Stream’s live environment model. - The company says the platform lets security teams build, deploy, and run AI agents that already understand their production environment. - StreamForce is available at stream.security/use-cases/force.

The details: - Stream Security says StreamForce removes the need to assemble infrastructure, normalize data, or pre-define use cases before deploying agents. - The company says its environment model updates in near real time as the environment changes. - StreamForce agents inherit that model and can identify what the environment looks like, what has changed, what is at risk, and what attack paths exist before a workflow runs. - Security teams can build an AI workflow in natural language and deploy it against a live production environment. - Stream says it spent three years building a normalized data fabric for AI. - The company describes that fabric as a live model that updates the moment anything changes. - Stream says that approach allows agents to act without burning tokens or hallucinating reality from raw logs. - CEO Or Shoshani compared the approach to giving AI agents a real-time map instead of making them infer the environment from a windshield view alone.

Between the lines: - StreamForce is aimed at a common security-automation pain point: teams often choose between building everything themselves or buying fixed workflows that do not fit their environment. - The launch suggests Stream Security wants to compete on context, not just automation. - The pitch also reflects a broader shift in security AI toward systems that understand infrastructure state before they generate recommendations or actions.

What's next: - Security teams can start using StreamForce now through general availability. - Hunt Consolidated’s early-access experience suggests Stream Security will point to customer validation as it expands adoption. - The key test will be whether security teams can use the platform to validate defenses and move faster without adding operational complexity.

The bottom line: - Stream Security is betting that AI in security works best when the agent starts with ground truth, not guesswork.