Features
- Claude Fable 5.2 — Anthropic's newest Fable model is available on direct Anthropic connections, with a 1M-token context window. Like the other Fable models, its adaptive thinking is always on: "off" uses low-effort thinking. Existing Anthropic connections pick it up on their next model refresh; Opus 5.5 stays the default. AWS Bedrock connections don't offer it yet.
Improvements
- Claude Agent SDK 0.3.321 — Claude sessions run on Claude Code 2.1.321 (from 2.1.316), which adds Claude Fable 5.2. A resumed session is less likely to start work before the app's hooks are in place, and reading a file a second time returns its content instead of an "unchanged" notice.
Bug Fixes
-
Automation docs match what runs — The automations guide the agent follows, and the automations page on the website, still described the removed
commandaction and listed agent events such asPreToolUseandStop, which never run actions. The agent could set up automations that silently did nothing. Both now list only the events that trigger automations and document webhook and script actions; scripts are the way to run your own code when an event fires. Fixes #1117. -
One broken automation no longer stops the rest — When a single automation in
automations.jsonwas invalid (for example one using the removedcommandaction, or a typo in a cron expression or regex), the whole file was ignored and none of your automations ran, while the Automations list still showed them all. Now only that automation is skipped: the list marks it Not running and its page says why. If the file isn't valid JSON, the list now says so instead of showing nothing, and automations that were already running keep running until it's fixed. The list now also shows script automations, which were missing from it. -
Automation prompts can't read the app's secrets — A prompt action expanded any environment variable, so
$CLAUDE_CODE_OAUTH_TOKENor$ANTHROPIC_API_KEYin a prompt, label or Telegram topic put the real value into the chat. Prompts now expand only the event'sCRAFT_*variables; anything else,$HOMEincluded, expands to nothing. Session names and labels also arrive as they are, without the backslashes that were added before quotes and$. -
Webhook bodies stay valid with quotes in values — A JSON body broke when a variable's value contained a quote or a line break (for example a session named
Fix "login"). Values are now inserted as proper JSON strings. A value is also no longer expanded a second time, so a session named$CRAFT_WH_SLACK_URLcan't pull your webhook secret into a request.