Unclear Which kind of error triggers the exit is not known, so it is unclear which failures still fall back to an empty conversation.
What
--continue and --resume reopen an earlier conversation. When starting one of these fails with a particular kind of error, Claude Code now stops with that error's own message instead of a generic failure.
- In the normal interactive mode, that error exits with its own message. Before, every failure was recorded as the same generic launch failure.
- In print mode (
claude -p), the error is printed and Claude Code exits with code 1. A failed resume of this kind is marked as aborted. Before, the error was logged and an empty conversation was returned. - A new check runs before the stored session is looked up.
Why
You see why a resume failed. Scripts using claude -p --continue or --resume now get a non-zero exit code they can detect, instead of silently carrying on with an empty conversation.
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Added since
A small documentation edit on Tools reference touched a line naming --continue after this was published.
In [non-interactive mode](/docs/en/headless) with `-p` and in the [Agent SDK](/docs/en/agent-sdk/overview), the limit applies per turn, not per session. A turn lasts from a message you send until Claude finishes responding. The limit reset…tools-reference see the edit
Confirmed since
Anthropic's documentation has since written up --continue, on Agent SDK reference - TypeScript.
Schedules a prompt to run on a 5-field cron schedule in local time. Set `recurring` to `false` to fire once at the next match. Jobs are session-scoped by default, and resuming with `--resume` or `--continue` restores jobs that haven't expi…agent-sdk/typescript see the edit
Confirmed since
Anthropic's documentation has since written up --resume, on Claude Code changelog.
* Fixed `claude --help` text for `--tools`, `--resume`, `--debug`, `--json-schema`, `--disable-slash-commands`, `--system-prompt-snapshot` and `claude update` to match actual behaviorchangelog see the edit
Added since
A small documentation edit on Manage sessions touched a line naming --resume after this was published.
* Default display name: interactive sessions you never name still get a default display name when they start. The default combines the working directory's name with a two-character suffix, for example `my-app-3f`, and identifies the sessio…sessions see the edit
Added since
A small documentation edit on Run Claude Code programmatically touched a line naming claude -p after this was published.
Calls to tools you didn't list run if they need no approval. In a [Manual mode](/docs/en/permission-modes#available-modes) `claude -p` run, a call that needs approval is denied unless something approves it, such as a [`PermissionRequest` h…headless see the edit
Confirmed since
Anthropic's documentation has since written up claude -p, on Configure permissions.
With Claude Code v2.1.302 or later, your app starts in `default` mode unless you set a permission mode or it gets one from [`permissions.defaultMode`](/docs/en/settings-reference#permissions-defaultmode) in its settings files. In `default`…agent-sdk/permissions see the edit
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubt
Which kind of error triggers the exit is not known, so it is unclear which failures still fall back to an empty conversation.
Anthropic's documentation agrees
Anthropic's documentation has since written up claude -p, on Configure permissions.
Anthropic's release notes agree
Changed claude -p and Agent SDK sessions with no permission mode to start in default mode on third-party providers, with telemetry off or on…