prompt
A Claude Code settings key, read out of the shipped bundle. It was in the build from v2.1.315 and was gone by v2.1.317.
Auto-mode is stricter: text you asked Claude to review can't grant permission on your behalf.
Not Anthropic's. This is the line from the earliest changelog entry here that named it, v2.1.221.
In the changelogs
33Releases whose published page names prompt.
-
v2.1.317
Hook settings now spell out the fields for agent and mcp_tool hooks
The description of valid hook settings now lists the fields for agent hooks and mcp_tool hooks
found in this entry's text
-
v2.1.317
Two internal name lists in Claude Code each gain an entry
A list holding command.describe, env.get and prompt.section now also holds tool.describe, and a list of input sources gains a mid-turn prompt entry
found in this entry's text
-
v2.1.312
Some prompt-rewriting hooks can now be skipped by a new check
The
prompt.section,agent.offerandattribution.texthooks now also skip when a new condition is metfound in this entry's text
-
v2.1.311
Settings can now hold prompt-type hooks as well as command hooks
Claude Code's hook settings reader now recognises prompt-type hooks, keeping their continue-on-block and status-message options
found in this entry's text
-
v2.1.304
Plugin mods are now checked for a screen-drawing hook
Claude Code now looks for a
ui.renderhook in plugin mods, and drafts written into the input box can carry decorationsfound in this entry's text
-
v2.1.299
Plugin hooks gain session identity and prompt media methods
Plugin hook modules can now call session.identity and prompt.media, mark registrations as withheld, and get fuller error details
found in this entry's text
-
v2.1.287
The prompt usage-data event is built differently
The usage-data event recorded when you send a prompt now gets its prompt fields from a different step, so its contents may change
found in this entry's text
-
v2.1.286
Plugins can now take over how the system prompt is put together
A new
prompt.composehook point lets a plugin module build the system prompt from its sections, and Claude Code falls back to its own if that failsfound in this entry's text
-
v2.1.286
Plugin hooks gain a prompt.compose method and an asUser option on submit
Plugin hooks can now call a new prompt.compose method, and prompt.submit now accepts an asUser option that must be true or false
found in this entry's text
-
v2.1.286
Several new internal log and event names
Claude Code gained a few new internal names for logging and events, tied to model lists, prompts, MCP images and Chrome tab groups
found in this entry's text
-
v2.1.281
Agent definitions may have an empty prompt
Agent definitions with an empty
promptare no longer rejectedfound in this entry's text
-
v2.1.280
New isTransparentWrapper() hook on an agent/tool adapter interface
Tool and agent adapters gain a new isTransparentWrapper() method, currently always false
found in this entry's text
-
v2.1.275
Skill/command execution now restricted to composer or bridge message origins
Skills and slash commands now only run when triggered from the composer or a bridge message
found in this entry's text
-
v2.1.274
Plugin SDK ($.) gains agent.register and session.refreshAgents, plus a system-prompt-delivered flag consumable once
Plugin SDK adds agent.register and session.refreshAgents, plus a one-shot flag for whether the system prompt was delivered
found in this entry's text
-
v2.1.274
SDK gains an agent.register call
SDK adds an agent.register call for defining named agents programmatically
found in this entry's text
-
v2.1.274
Draft restoration now preserves pasted-content mode correctly
Restoring a saved draft now correctly preserves pasted content only for prompt-mode drafts
found in this entry's text
-
v2.1.271
Slash-command resolution helpers reworked for poll-event/prompt modes
Slash-command resolution logic reworked to distinguish bash, poll-event, and prompt modes
found in this entry's text
-
v2.1.269
WebFetch/artifact-read tool also moves to
create()instance pattern with new renderToolUseMessage and prompt hookThe WebFetch tool moves to the newer
create()internal structure and gains a verbose-mode message and prompt hookfound in this entry's text
-
v2.1.268
Permission decisions now record how they were decided (
decidedBy)Permission approvals now record whether a hook, the host tool, or a normal prompt approved them
found in this entry's text
-
v2.1.267
Compaction can now be intercepted and rewritten by a session.compact plugin hook
A new
session.compactplugin hook can intercept and rewrite conversation compactionfound in this entry's text
-
v2.1.265
ui.invalidate now supports invalidating command.describe
The ui.invalidate plugin operation can now also invalidate command.describe
found in this entry's text
-
v2.1.259
New functionHooks event: prompt.context
A new prompt.context hook event lets plugins add named context blocks to a prompt, capped at 32 blocks and 32000 characters.
found in this entry's text
-
v2.1.259
Plugin UI hooks can now invalidate on prompt.context, not just prompt.section
Plugins can now invalidate prompt.context specifically through ui.invalidate, alongside ui.render and prompt.section.
found in this entry's text
-
v2.1.257
Queued prompt latency now tracked
Queued prompts now carry submit and dequeue timestamps so queue latency is tracked.
found in this entry's text
-
v2.1.257
AskUserQuestion-style tool prompt gains a leanPrompt parameter
The AskUserQuestion prompt now receives a lean-prompt flag that can suppress a guidance snippet.
found in this entry's text
-
v2.1.242
Plugin modules with a
$API that can hook events and redraw the terminal UIPlugins can install code that intercepts tool calls and repaints parts of the terminal interface.
found in this entry's text
-
v2.1.239
Artifact tool gains a
readaction, with multi-file browsing and delete built but gated offClaude can now read back a published artifact's contents; browsing, multi-file publishing and deletion are built but off.
found in this entry's text
-
v2.1.221
Auto-mode security classifier gains three consent boundaries
Auto-mode is stricter: text you asked Claude to review can't grant permission on your behalf.
found in this entry's text
-
v2.1.221
Whiteboard hardening: permission verdicts, state sanitising, merge-state validation
Whiteboard permission checks and board data validation got stricter, but none of it runs here.
found in this entry's text
-
v2.1.202
/loopExplicit Stop viaScheduleWakeup({stop: true})found in this entry's text
-
v2.1.196
Prompt ID in Hook Events
found in this entry's text
-
v2.1.186
Shell Variable Tracking: More zsh Variables Covered
found in this entry's text
-
v2.1.30
Enhanced Custom Subagent Schema
found in this entry's text
First cited
0No release's published evidence quotes this name. The ledger indexes the Evidence lines of every changelog on this site, so this says nobody here has ever quoted it; it says nothing about how old the name is.
Presence across releases
2Build by build
4One row per release since the first build this name was read out of. Absent means the build was read and the name was not in it, never mined means no bundle for that release was ever archived, and a declared type or a default is only ever what that release's own bundle stated.
Read out of the published npm bundle release by release, and out of Anthropic's own documentation as this site captured it. Nothing on this page is a description anybody here wrote about what the settings key does. All settings keys.