What
Hooks are your own scripts that Claude Code runs at set moments, such as UserPromptSubmit when a prompt is submitted, or UserPromptExpansion. Some slash commands run in a fork, a separate copy of the conversation. When a hook blocks one of these, Claude Code used to add "Original prompt:" with your text unless the hook set suppressOriginalPrompt.
The echo is now also left out when:
- the hook set
suppressOriginalPrompt - the text you typed differs from the text the hooks saw
- the hook's state is unknown or has changed
Blocked results are marked as dropped, and a run that was cut short is flagged as such.
Why
A prompt your hook blocked is less likely to show up again in the transcript.
Documented inclaude-code/agent-sdk/typescript
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.
Confirmed since
Anthropic's documentation has since written up UserPromptExpansion, on Intercept and control agent behavior with hooks.
* `UserPromptSubmit` and [`UserPromptExpansion`](/docs/en/hooks#userpromptexpansion): Claude Code blocks the prompt with a message naming the hook and the timeout, and the session continues. Before v2.1.208, Claude Code ended the query wit…agent-sdk/hooks see the edit
Confirmed since
Anthropic's documentation has since written up UserPromptSubmit, on Intercept and control agent behavior with hooks.
* `UserPromptSubmit` and [`UserPromptExpansion`](/docs/en/hooks#userpromptexpansion): Claude Code blocks the prompt with a message naming the hook and the timeout, and the session continues. Before v2.1.208, Claude Code ended the query wit…agent-sdk/hooks see the edit
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agrees
Anthropic's documentation has since written up UserPromptSubmit, on Intercept and control agent behavior with hooks.