July 2026

Everything that changed in NoClick in July 2026.

Improvements and fixes2
  • New agent nodes now default to the gpt-5.6-luna model instead of a mini model. Agents you have already configured are unchanged.

  • The model picker labels models the provider serves at no cost with a Free tag, sorts them above paid models, and finds them when you search for "free" even when the model name does not contain the word. The five CLI coding agents carry a CLI Agent tag so they are not mistaken for single models.

Improvements and fixes1
  • When a model provider rejects an agent run, the error now reads as a sentence and offers the fix. A rejected key gets an Open credentials button, an account with no API credits gets a link to that provider's billing page, and a plan cap or a provider outage says which one it is. Each message states that the balance in question is the provider's, not your NoClick credits, and the provider's original text is kept below it.

Improvements and fixes1
  • Agent nodes running on a CLI coding agent (Claude Code, Codex, OpenCode, OpenClaw, or Hermes) now tell you when no key or AI subscription is connected. The node is marked incomplete on the canvas and its config panel says credentials are required. OpenCode and Hermes gave no warning at all before, so the first sign of the problem was a failed run.

A smoother agent chat

  • When an agent run stops, the error banner now explains what went wrong and offers one-click fixes. Add credential opens a popup to connect or select credentials, and Change model opens the full model picker with search, provider filters, and favorites.
  • Clicking the model name at the top of the chat opens that same model picker, the one the canvas agent node uses.
  • A run that fails no longer shows leftover output from an earlier run, and a stuck run now clears with a clear error instead of spinning.

Agents are ready to chat right away

New AI agents now show in the workflow's Interface tab by default, so you can chat with one the moment it is built. After the AI builder finishes a workflow with a chat agent, a one-time walkthrough points you to the Interface tab to try it. You can still turn Show in Interface off for background agents, and existing workflows are unchanged.

Faster CLI coding agents

The CLI coding agents (Claude Code, Codex, OpenCode, OpenClaw, and Hermes) now keep a warm sandbox between turns instead of starting fresh each time, so replies come back faster and one agent can hold several conversations at once. A small indicator on the agent node shows how many sandboxes it is currently holding, with a live note while a turn is running.

Sandbox time now costs credits

CLI coding agents run in a cloud sandbox that NoClick provides, and that sandbox time is now charged at 0.05 credits per minute of uptime, billed per turn. This applies even when the agent runs on your own API key or AI subscription, since the sandbox is NoClick's compute. The built-in LLM agent runs in-process and has no sandbox charge.