Caudra 0.2 Preview 8
- Claude Haiku 5.5 is the Anthropic Fast model.
- Caudra speaks the current MCP protocol, and servers can ask you questions, report progress, and share resources.
- The experimental decision engine reranks memory notes, tool search, retained tool output, and code context, the transcript shows what each decision changed, and workflows gain
rerank(). - Favorite and 10 recent models in
/model, and tasks run on the model you name. - Forks open in a new Herdr or Orca tab, and tabs are named after the session.
- New sessions can open in build mode, and
/sessionscan delete the current session. - Fixes for list scrolling, thinking levels, session moves, Orca's bell, and
caudra updateoutput.
Important
Close every other Caudra session before you start Preview 8. On its first start, Preview 8 upgrades the session database in your state directory, and the upgrade needs exclusive access. This includes other terminals, editor sessions over ACP, and
caudra storagecommands. While any of them has the database open, Preview 8 lists the sessions to close and does not start, and nothing is changed.The upgrade keeps your sessions and saves the old database as
caudra.db.v24.baknext tocaudra.db. Preview 7 and earlier cannot open the upgraded database, andcaudra rollbackrestores only the previous binary, not the database.
Note
If you raised
decisions.timeout_msfor a slow decision engine and usetool_search, setrerank_timeout_msas well. Tool search now uses that deadline, which defaults to 2 seconds.
Models
Section titled “Models”- Claude Haiku 5.5 is the Anthropic Fast model.
anthropic/claude-haiku-5-5costs $0.10 per million input tokens and $0.50 per million output tokens, with cache reads at $0.01. Once a prompt passes 100k tokens, cached tokens included, the whole request costs five times those rates. It runs at a 372k working window, andclaude-haiku-5-5-1mopens the full 1M window. Copilot lists it asclaude-haiku-5.5. It replacesclaude-haiku-4-5as the Fast model, so on an Anthropic session, session titles, completion-goal checks, and other Fast work run on it unless you bound those jobs to a model. Bind Fast toanthropic/claude-haiku-4-5in/modelto keep the older model. See Anthropic and Model jobs. - Sonnet 5.5 cache reads cost less. Caudra now counts Claude Sonnet 5.5 cache reads at the new price of $0.10 per million tokens, on Anthropic and Copilot.
- Thinking levels a model rejects are no longer sent.
minimalnow sends the shallowest level a model has, where it used to sendnoneand turn reasoning off. On Copilot, thinkingoffon a model that always reasons, such asgpt-6.1-sol, runs at its lowest level instead of failing withUnsupported value: 'none' is not supported. Anonesetting now means the same asoffon every model. A customopenaioropenai-responsesmodel withoutreasoning_optionstakes no thinking setting. See Model defaults.
Model picker and tasks
Section titled “Model picker and tasks”- Favorite models. In
/model, pressCtrl+Fon a model to mark or unmark it as a favorite. Favorites are listed first and kept across sessions, and/favorite-nextorCtrl+X cswitches to your next favorite. Recent now keeps the last 10 models instead of 4, without your favorites. Every model also keeps its row in its provider's section, marked★for a favorite and↺for a recent model. See Model jobs. - Run a task on a model you name. Ask for a model, as in "ask opus and sol to review this change", and the main agent starts one task per model you name. The
tasktool takes amodel, either an exactprovider/modelname or a short name such asopusthat resolves to the newest match, and athinkinglevel fromofftomax. Both win over a profile'ssubagent_modelandsubagent_thinking. A name that fits several models fails and lists them. A task's result, footer, and card name the model and level it actually runs on.[agent] task_model_choice = falseturns this off. See Models and reasoning.
Decision engine
Section titled “Decision engine”- Reranking features. Six new experimental decision features use the engine as a reranker, and each is off by default.
memory_searchlists relevant memory matches first and names the ones it left out.memory_recallreminds the agent of up to three notes that may matter for your new message.memory_duplicatespoints out an existing note a new one may repeat.compaction_retentionkeeps the tool results that matter to the current goal when a session compacts, instead of only the newest three.output_pointerspoints a truncated tool result at the omitted lines most likely to matter.code_contextmoves the most relevant rows of acode_contextresult to the top. When the engine fails or times out, each behaves as if it were off. They use a newrerank_timeout_msdeadline (2 seconds) andthresholds.relevance_floor(0.1). Withtool_searchenforced, tool search now scores and reorders its first 16 keyword matches instead of picking one, and never drops a match. See Reranking. - Workflows can rerank.
rerank(query, candidates)returns#{ index, score }entries, most relevant first, with the optionsmodel,timeout_ms,threshold, andtop_k. Each batch of 16 candidates is one journaled decision, so a resumed run asks only for the rest. In/workflow,decide()andrerank()calls are labelled with their question IDs or candidate counts. Adecide()orrerank()timeout may now reach the larger oftimeout_msandrerank_timeout_ms. See Reranking candidates. - Decision rows in the transcript. When a decision feature changes what happens, the transcript shows a row that names the feature and what it did, such as
code_context · moved 3 of 12 rows up. A row also appears when the engine fails and the default applies, and a dimshadowrow shows what a feature in shadow mode would have done. Click a row to see the candidates and scores behind it. Rows are saved with the session, the model never sees them, and[ui] show_decisions = falsehides them. See Decision rows. - Shell duration estimates read plainly. A shell card now shows an estimate only for a command expected to take more than a second, in the units of its timeout, such as
ETA 10s–2m.
- The current MCP protocol. Caudra now speaks MCP
2026-07-28and still speaks the older protocol that starts withinitialize, so servers built on either work. When a server announces that its tools, prompts, or resources changed, Caudra lists them again. Caudra never offers roots, sampling, or logging, and writes server log messages tocaudra.logat debug level. See Protocol versions. - Servers can ask you questions. During a tool call, a server can ask for input with a form of typed fields or a link to open, such as a sign-in page. The question form checks each answer and lets you review before sending.
F3declines andEsccancels. A link opens in your browser only after you agree. Autopilot may answer a form from the main agent's call, but never a link, a form from Executor, or a form asking for a credential. ACP clients that support MCP elicitation get the question too. See Server questions. - Long calls report progress. A server's progress shows beside the card's label, and each report restarts the
timeoutwait, up to ten timestimeout. Stopping the run tells the server the call was cancelled. After a dropped connection, a call that is not safe to repeat fails withconnection lost; the call's outcome is unknown. See Long calls. - Resources and completions. Mention a server's resource as
@server:uri, and Caudra reads it into your message like a mentioned file. Typing@lists resources beside your files, and template values and prompt arguments complete from the server. The model can list and read resources with the newmcp_resourcetool, which works in plan mode and asks first. A prompt's images and embedded resources now reach the model. See Resources. - Server instructions reach the model. The instructions each running server gives for its tools are added to the system prompt, up to 4096 bytes per server. See Server instructions.
- Executor and Home Assistant are recognised. Executor gets a card that shows how a run ended, its logs, and what a paused run waits for. An Executor
resumecall that approves a paused run always asks you, even under autopilot. Add?elicitation_mode=nativeto Executor's URL so it asks you directly. Home Assistant gets cards for intents and its live context, and a call that controls alock,alarm_control_panel,cover, orvalvealways asks, even with an allow rule or in Auto. See Recognised servers. - Read-only tools work in plan mode. Tools that Executor or Home Assistant mark read-only, such as Home Assistant's
GetLiveContext, can be called in plan mode and by read-only agents. They still ask, with an option to always allow the tool. See MCP tool calls. - MCP cards open whole. A finished MCP call's card shows its arguments and its whole result instead of the first three rows. Compact view and
ui.always_collapsedstill fold it. A card that ends with aclick for rawrow switches to the result's raw JSON and text. A result the server marks as an error draws as a failed card. See Results in the transcript. /mcpshows what a server reported. Select a running server to see its name and version, the protocol version, the features it offers, and the first lines of its instructions. See Runtime toggling.- A sign-in the server refuses is reported.
caudra mcp authnow checks that the server accepts the stored sign-in, and signs in again when it does not. When a fresh sign-in is still refused, as on Home Assistant's/api/mcpfor a user who is not an administrator,/mcpshows the server as Refused instead of asking you to sign in again. Home Assistant's/api/mcp/assistworks for any user. See OAuth.
Interface fixes
Section titled “Interface fixes”- The wheel scrolls lists. In
/decisions,/automations,/workflow,/peersand/messages, the wheel over the list now scrolls it instead of moving the selection. - Files picked with
Ctrl+Sare mentioned. The file picker now inserts the path as an@mention, so the file is attached when you send the message. See Mention a file with @. - The
Ctrl+Xpanel takes the mouse. Click a chord in the panel to run it, or click anywhere else to cancel. - Hidden Orca tabs ring when a turn ends. A hidden Caudra tab in an Orca remote workspace could show
workinglong after its turn ended. Caudra now rings the bell when a turn ends or a prompt waits while its terminal does not have focus, so Orca marks the tab unread, and sends its status again when you return. Turn on Terminal Bell in Orca's notification settings for a desktop notice as well.notifications = "off"stops the bell. See Orca. - Prompts, popups and scrollbars follow the theme. The permission prompt, the
@and#completion lists, the sandbox manager, and scrollbars now use the theme's colours, so a dark theme on a light terminal no longer shows light patches.
Sessions
Section titled “Sessions”- Open new sessions in build mode. Set
default_mode = "build"under[ui]to start new terminal UI sessions in build mode instead of plan mode. A resumed session still reopens in the mode it was left in. Seeui. - Delete the current session. In
/sessions,Ctrl+Don the current session now deletes it after the usual confirmation and opens a new session in its place, in the same mode and directory. A session that is still working cannot be deleted until it is idle. See Sessions. - Tabs are named after the session. Caudra sets the terminal window title to the focused session's title, so most terminals, Orca, and Herdr label the tab with it. A tab you named yourself keeps its name. See Tab names.
- Forks open in a new Herdr or Orca tab. Inside a Herdr pane or an Orca terminal, Fork here resumes the new session in a new tab with a Caudra of its own, and the session you forked from stays open. Remote, sandbox, and ephemeral sessions still fork into a tab of the current Caudra. See Forks in Herdr and Orca.
- Memory writes wait out a busy database. The
memorytool failed withdatabase is lockedwhile another Caudra process held the database for more than 5 seconds. Memory writes now retry for up to 30 seconds, like session saves do. - Pending reverts no longer block moves.
/move-session,/migrate-sessions, and/worktree removewere refused or failed to start when a session already in the destination had a pending revert. Sessions in the destination no longer affect a move.
Command line
Section titled “Command line”caudra updateno longer prints its installer. It names the release it moves to, installs it without asking, and ends with thecaudra rollbackcommand that returns to your current version.caudra update --show-scriptprints the checksum-verified installer and asks before running it, as before.-yand--no-colorare still accepted, so existing scripts keep working. Seecaudra update.