Velqa Sandbox Agent — run code in an ephemeral Docker sandbox (BETA)

Velqa's Sandbox Agent lets you build AI agents that actually run code inside an isolated, ephemeral Docker sandbox, straight from the dashboard. This is a BETA feature, gated by entitlement: if it isn't enabled on your account, the dashboard shows an upgrade prompt instead.

The model: one ephemeral sandbox per turn

Unlike plain chat, where the model only produces text, a Sandbox Agent gets a real execution environment:

• On every conversation turn, Velqa spins up a fresh Docker sandbox (Python or Node, depending on the agent).
• Inside it, the agent can write and read files, run shell commands, execute code, search the web, and more (see agent tools).
• The sandbox is isolated and ephemeral: it expires after roughly 10 minutes, then disappears. Nothing persists from one sandbox to the next except what you keep via checkpoints or publish as an artifact.

BETA and access

• The feature is entitlement-gated during the closed beta. Without access, you see an upgrade prompt instead of the Sandbox UI.
• Billing is based on per-minute compute of the sandbox. During the beta this is shadow billing: usage is measured, but the impact on your balance is indicative only.

When to use it instead of plain chat

Reach for the Sandbox Agent when you need the AI to do something, not just describe it:

• run and verify code (Python/Node), run tests, manipulate files;
• process a file you upload into the workspace;
• chain several tool-driven steps (search, compute, produce a downloadable deliverable).

For simple Q&A or text generation, stay on chat: it's faster and cheaper.

Next steps

• Create and configure an agent — name, runtime, model, instructions.
• Conversations and turns — run a turn, streaming, expiry, stopping.
• Agent tools — what the agent can do inside the sandbox.
• Workspace and artifacts — upload files, publish deliverables.