tancheng33/dsh-code-runtime-container
tancheng33★ 0TypeScriptLast synced: 2026-08-16
Container-isolated backend for the DeepSeek Harness code-execution seam: Code Mode programs run in a fresh container with no network, a read-only rootfs, and kernel-enforced memory, CPU, and pid ceilings
README excerpt
dsh-code-runtime-container English 中文 A container-isolated backend for the DeepSeek Harness code-execution seam ( ctx.codeRuntime ). Code Mode programs run in a fresh container with no network, a read-only root filesystem, every capability dropped, and kernel-enforced memory, CPU, and pid ceilings. Why The seam declares three well-known isolation substrates and ships one. From @deepseek-ai/dsh-code-runtime 's own README: Only the worker-thread backend ships — 'process' / 'container' are declared well-known isolation values with no implementation; a hard security boundary awaits a container backend. And the shipped backend is equally explicit about its posture: Containment, not a security boundary : trust posture is bash-equivalent by design. That is a reasonable default — Code Mode programs are model-written code, and so is everything bash runs. But it means a run code program executes inside the agent's own process , with the agent's network access, the agent's filesystem, and whatever the agent's environment holds. This backend is for deployments where that is not acceptable. worker-thread (shipped) this backend Substrate a Worker in the agent process a fresh container per run Ne…
View full README on GitHub →