wdx prompt
Answer the prompts the project’s Claude Code agents wait on.
wdx prompt [--project-id <id> | --project-name <name>] [flags]Lists the agents of the project with a prompt open, in the agents’ order, asks which one, shows its question, detail and options, and sends the answer through the agent’s holder, which presses the keys one at a time and only while the screen still shows the prompt you saw.
A number picks a choice. A checkbox list takes the numbers to tick, as 1 3, and asks before sending. A text option asks for the text. Chat about this declines the prompt, and the agent goes on in its own tab or wdx agent attach. A decision that asks several questions asks each in turn: Claude Code moves on to the next as each is answered, nothing is submitted, and after the last the answers are shown for review and sent only on an explicit yes. A prompt that changed meanwhile is said, and the list read again.
Only Claude Code’s prompts are read, so only its agents are listed. --json lists them and asks nothing; with --no-input or no terminal the command needs --json. Without a project flag, the project the working directory is in.
| Flag | Default | Description |
|---|---|---|
--node <name> |
run the command on the node of this name, another of your account, through a session | |
--project-id <id> |
id of the project (default: the one the working directory is in) | |
--project-name <name> |
name of the project (default: the one the working directory is in) |
- at most one of
--project-idor--project-name
Examples
Section titled “Examples”# Answer a prompt of the project you are inwdx prompt
# Those of another projectwdx prompt --project-name app
# List them as JSON, asking nothingwdx prompt --jsonOutput
Section titled “Output”What the dialogue asks goes to stderr; stdout holds the answered agent, or no prompts when none waits. With --json: {“agents”: […]}, the agents with a prompt, each with its “prompt” as the API serves it, children not nested
With --json, result is one of: answered.
Errors
Section titled “Errors”| Code | Exit | When |
|---|---|---|
aca_unsupported |
1 | the agent’s AI coding agent cannot do this yet |
agent_not_running |
1 | the agent’s AI coding agent is not running |
agent_removed |
1 | the agent was removed and stays only for its children |
answer_failed |
3 | the agent’s holder could not answer the prompt, or did not see the answer taken |
cancelled |
1 | the question was answered no |
daemon_unavailable |
3 | the daemon is not running or not answering |
not_found |
1 | nothing has this name or id |
prompt_changed |
1 | the agent’s prompt is not the one answered, or changed while it was answered |
usage |
2 | the request, command line or a setting is invalid |
internal |
3 | an unexpected failure: database, filesystem or git |
With --node, also: code_not_set_up, code_wrong, connection_lost, daemon_unavailable, hub_unreachable, hub_version_unsupported, node_key_mismatch, node_not_found, node_offline, node_version_unsupported, not_logged_in, remote_access_off, remote_not_for_agents, route_not_allowed, session_closed, session_required, too_many_streams.