Nasiko adds TokenOps budget controls for AI agents to its OpenRuntime platform
In brief
- Nasiko announced TokenOps on November 16, 2026, within its OpenRuntime platform.
- Per-agent budgets set by operators trigger alerts at 80% and pause agents at the limit.
- Cost breakdowns span agent, team and model, across providers including OpenAI and Gemini.
- Token spend could fall by up to 50% through routing, per Nasiko's unverified estimate.
How TokenOps works
TokenOps isn't a standalone product. It lives inside OpenRuntime, which Nasiko describes as a control plane for AI agents (it runs as a single server process that proxies the calls those agents make). That design matters. Because requests flow through it, the platform can attribute costs across multiple providers, including OpenAI and Gemini.
The feature breaks down token usage and associated costs by individual agent, by team, and by model. Nasiko frames the problem around what it calls the "agent estate" (the growing collection of AI agents that teams run across different frameworks). According to the project's materials, no SDK migration is needed, so existing agents and workflows can keep running while teams adopt centralized budget governance.
The 50% routing claim
On the routing side, Crypto Briefing reported potential token spend reductions of up to 50% through effective routing. The report didn't say who made that estimate, and it's best read as Nasiko's own projection rather than a measured result.
It's a projection, not a verified result.
Deployment and history
The open-source runtime bundles other tools alongside the budget controls. It supports coding agents including Claude Code and Codex, and custom agents can be deployed in a containerized environment using Docker. Operators don't have to run it themselves: they can self-host the platform or use a hosted option. The project is publicly available on GitHub.
Nasiko introduced its core control plane in May 2026, according to Crypto Briefing's account of the company's materials. In September, it onboarded onto the DigitalOcean Marketplace, giving the platform another distribution channel.
TokenOps arrives six months after that control plane launch.
Frequently asked questions
How does Nasiko's TokenOps track AI agent spending?
TokenOps runs inside OpenRuntime, a single server process that proxies the calls AI agents make. Because requests flow through it, the platform can attribute costs across providers including OpenAI and Gemini, and it breaks token usage down by agent, by team, and by model.
What happens when an AI agent hits its TokenOps budget?
Operators set a budget for individual agents. The system sends an alert when spending reaches 80% of the configured limit, and TokenOps pauses the agent once it hits the limit.


