Submilli is a code-execution runtime with a semantic permission model for business agents that generate code. A support agent might be allowed to issue a refund of up to $500 for platinum clients and $100 for everyone else; putting that rule in the prompt only works some of the time, so the runtime enforces it outside the model's control. It ships as a container image and a Helm chart for self-hosting, and is Apache 2.0 licensed.
Be the first to rate Submilli Runtime. Your feedback helps other users decide.
Declare rules like a refund ceiling per customer tier in advance rather than in a prompt.
Execute agent-written code through the runtime so the permissions are enforced by the runtime itself.
Use the published container image and Helm chart in your own cluster.
Similar tools in the same category — compare side-by-side
Orbit is a local-first runtime for coding agents: you keep using your authenticated agent CLI and Or...
Otōto is an MCP server for coding agents — Claude Code, OpenCode, or anything else that speaks MCP —...
Gutsy runs local decision models that return calibrated probabilities. You send a state, text or JSO...
Guides and comparisons about code tools
Nine launches that draw the line around an agent: policy proxies, guard hooks, semantic permissions, sandboxes, worktrees and a proof checker.
Which AI coding assistant actually saves you time? GitHub Copilot vs Cursor vs Codium in 2026, with real comparison and recommendations by use case.
How to choose an AI coding tool in 2026: inline completion, repo chat, and local privacy, compared across Copilot, Cursor, Codeium, and more.
Explore the best AI code tools for web development in 2026. Accelerate frontend and backend development with AI-powered coding assistants.
Discover the best AI code tools for testing automation in 2026. Generate tests, find bugs, and automate QA workflows with AI-powered testing solutions.