Skip to content

Caudra 0.2 Preview 6

View Markdown~1,699 tokens

v0.2.0-preview.6 on GitHub

  • Autopilot (experimental) answers permission prompts, questions, and ready plans while a session works toward its goal.
  • Chosen environment variables reach shell commands.
  • MCP results are drawn readably, and MCP servers without client registration can sign in.
  • What is new shows after an update.
  • Orca terminal support.
  • Enter answers a permission prompt in terminals where it did not.

Important

Close every other Caudra session before you start Preview 6. On its first start, Preview 6 upgrades the session database in your state directory, and the upgrade needs exclusive access. This includes other terminals, editor sessions over ACP, and caudra storage commands. While any of them still has the database open, Preview 6 does not start. It lists the open sessions to close instead, and nothing is changed.

The upgrade keeps your sessions and saves a copy of the database from before it as caudra.db.v23.bak next to caudra.db. Preview 5 and earlier cannot open the upgraded database, and caudra rollback restores only the previous binary, not the database.

  • Keep a session moving. With experimental.autopilot on, /autopilot <goal> sets a goal and hands what would wait for you to a separate delegate model: permission prompts, the question tool, questions asked in prose, a plan that is ready, and the stalls that end a run, such as a goal loop cap or a provider outage. Autopilot ends when the goal is met or impossible, a budget runs out, or you stop it. Decisions use the new Autopilot model job, which follows Best unless you bind it in /model or with /autopilot model. See Autopilot.
  • Limits. The delegate approves a permission prompt once and never adds a rule. It never approves prompts that plan mode raises, tools that always prompt, protected files and commands, or any prompt in a session that content screening flagged. Each engagement may start 100 turns without you by default, and max_cost and max_minutes under [autopilot] add spend and time limits.
  • You stay in charge. Every decision is a transcript row, and a footer chip opens a panel with the budget and recent decisions. A key or click on a card autopilot is deciding makes it yours. Esc Esc, Ctrl-C while the agent works, or /autopilot off stops autopilot. An item it leaves to you raises an urgent notification.
  • Headless and CI. caudra -p --autopilot "/goal <condition>" runs unattended, and a prompt autopilot leaves to a human is denied. SDK clients engage it with --autopilot or the autopilot_start control. See Headless Mode.
  • Forward the variables your commands need. /env lists every variable Caudra can see, shows which ones reach shell commands, and edits [agent.shell_env] in the global caudra.toml. pass forwards names such as DOCKER_HOST or AWS_*, and set gives commands their own values, with {env:NAME} for secrets. A project .env never supplies a forwarded value, and a project config cannot hold the table. See Forwarding more variables.
  • Sandboxes forward their guest's variables. A sandbox profile's shell_env chooses which guest variables reach commands, all of them by default. Values stay in the guest. See Shell environment.
  • Readable results. Each block of an MCP result is drawn by its type: JSON is indented and coloured, other text is rendered as markdown, and images, audio, and resources become labelled rows with their type and size. The model receives the same text as before.
  • Cleaner header. A card reads server › tool with short arguments beside it, followed by the result's size and an estimate of the tokens the model reads, such as JSON · 128 lines · ~3.2k tokens.
  • Code it ran. When a call runs code, such as a code, script, or sql argument, the card shows it numbered and highlighted above the result. See Results in the transcript.
  • Servers without client registration. Caudra now logs in to servers that accept OAuth client ID metadata documents, such as Home Assistant, with no client to configure. These logins used to fail with "no stored client and server has no registration endpoint". The server must be able to reach caudra.ai, where Caudra's document lives. See OAuth.
  • Stricter discovery. Login discovery follows the MCP spec more closely, and servers without S256 PKCE are refused. A client is reused only with the authorization server that issued it, so the next interactive login after this upgrade gets a new client once. Stored tokens keep refreshing.
  • Enter answers a fresh prompt. In Herdr and some other terminals, Enter did nothing on a new permission prompt until you clicked or pressed another key. A key you were already holding when the prompt appeared is still ignored, and the footer says whether to let go or to wait. See Permission prompts.
  • Tooltips. Resting the mouse on any part of a permission prompt explains what it does and which key presses it.
  • What's new after an update. The first time the terminal UI starts after an update, it shows the notes for every release since the one you last ran. /changelog shows them again, and /changelog all shows every release. The notes ship inside Caudra, so they work offline. Set ui.changelog_on_update = false to show them only through /changelog. Every release's notes are also on the Release notes page and in /docs.
  • Orca terminals. Inside an Orca terminal, Caudra reports to Orca its state, its model, and the label of a prompt that waits for you. /worktree creates and opens worktrees through Orca for repositories Orca knows, and git otherwise. Shell commands get Orca's terminal variables, and an orca skill teaches the model Orca's CLI. See Orca notifications and Orca worktrees.
  • More accurate pane status. Herdr and OSC 7501 pane status now match what Caudra is doing more closely, and automation notify() messages always arrive as notifications, even inside Herdr and Orca.
  • To-do list of the task you view. The sticky to-do list follows the conversation you view, so a task shows its own list or none.

Website privacy and analytics