Detailed changes
@@ -9,7 +9,7 @@ use std::{any::Any, path::PathBuf};
use anyhow::{Context as _, Result};
use gpui::{App, AppContext as _, SharedString, Task};
-use project::agent_server_store::{AllAgentServersSettings, CLAUDE_CODE_NAME};
+use project::agent_server_store::{AllAgentServersSettings, CLAUDE_AGENT_NAME};
use crate::{AgentServer, AgentServerDelegate, load_proxy_env};
use acp_thread::AgentConnection;
@@ -24,7 +24,7 @@ pub struct AgentServerLoginCommand {
impl AgentServer for ClaudeCode {
fn name(&self) -> SharedString {
- "Claude Code".into()
+ "Claude Agent".into()
}
fn logo(&self) -> ui::IconName {
@@ -232,8 +232,8 @@ impl AgentServer for ClaudeCode {
let (command, root_dir, login) = store
.update(cx, |store, cx| {
let agent = store
- .get_external_agent(&CLAUDE_CODE_NAME.into())
- .context("Claude Code is not registered")?;
+ .get_external_agent(&CLAUDE_AGENT_NAME.into())
+ .context("Claude Agent is not registered")?;
anyhow::Ok(agent.get_command(
root_dir.as_deref(),
extra_env,
@@ -1688,7 +1688,7 @@ mod tests {
None,
prompt_capabilities.clone(),
available_commands.clone(),
- "Claude Code".into(),
+ "Claude Agent".into(),
"Test",
EditorMode::AutoHeight {
min_lines: 1,
@@ -1713,7 +1713,7 @@ mod tests {
// Should fail because available_commands is empty (no commands supported)
assert!(contents_result.is_err());
let error_message = contents_result.unwrap_err().to_string();
- assert!(error_message.contains("not supported by Claude Code"));
+ assert!(error_message.contains("not supported by Claude Agent"));
assert!(error_message.contains("Available commands: none"));
// Now simulate Claude providing its list of available commands (which doesn't include file)
@@ -1730,7 +1730,7 @@ mod tests {
assert!(contents_result.is_err());
let error_message = contents_result.unwrap_err().to_string();
- assert!(error_message.contains("not supported by Claude Code"));
+ assert!(error_message.contains("not supported by Claude Agent"));
assert!(error_message.contains("/file"));
assert!(error_message.contains("Available commands: /help"));
@@ -2386,7 +2386,7 @@ impl AcpServerView {
fn current_model_name(&self, cx: &App) -> SharedString {
// For native agent (Zed Agent), use the specific model name (e.g., "Claude 3.5 Sonnet")
- // For ACP agents, use the agent name (e.g., "Claude Code", "Gemini CLI")
+ // For ACP agents, use the agent name (e.g., "Claude Agent", "Gemini CLI")
// This provides better clarity about what refused the request
if self.as_native_connection(cx).is_some() {
self.active_thread()
@@ -2395,7 +2395,7 @@ impl AcpServerView {
.map(|model| model.name.clone())
.unwrap_or_else(|| SharedString::from("The model"))
} else {
- // ACP agent - use the agent name (e.g., "Claude Code", "Gemini CLI")
+ // ACP agent - use the agent name (e.g., "Claude Agent", "Gemini CLI")
self.agent.name()
}
}
@@ -6749,7 +6749,7 @@ impl AcpThreadView {
fn current_model_name(&self, cx: &App) -> SharedString {
// For native agent (Zed Agent), use the specific model name (e.g., "Claude 3.5 Sonnet")
- // For ACP agents, use the agent name (e.g., "Claude Code", "Gemini CLI")
+ // For ACP agents, use the agent name (e.g., "Claude Agent", "Gemini CLI")
// This provides better clarity about what refused the request
if self.as_native_connection(cx).is_some() {
self.model_selector
@@ -6758,7 +6758,7 @@ impl AcpThreadView {
.map(|model| model.name.clone())
.unwrap_or_else(|| SharedString::from("The model"))
} else {
- // ACP agent - use the agent name (e.g., "Claude Code", "Gemini CLI")
+ // ACP agent - use the agent name (e.g., "Claude Agent", "Gemini CLI")
self.agent_name.clone()
}
}
@@ -29,7 +29,7 @@ use language_models::AllLanguageModelSettings;
use notifications::status_toast::{StatusToast, ToastIcon};
use project::{
agent_server_store::{
- AgentServerStore, CLAUDE_CODE_NAME, CODEX_NAME, ExternalAgentServerName,
+ AgentServerStore, CLAUDE_AGENT_NAME, CODEX_NAME, ExternalAgentServerName,
ExternalAgentSource, GEMINI_NAME,
},
context_server_store::{ContextServerConfiguration, ContextServerStatus, ContextServerStore},
@@ -942,7 +942,7 @@ impl AgentConfiguration {
let user_defined_agents = agent_server_store
.external_agents()
.filter(|name| {
- name.0 != GEMINI_NAME && name.0 != CLAUDE_CODE_NAME && name.0 != CODEX_NAME
+ name.0 != GEMINI_NAME && name.0 != CLAUDE_AGENT_NAME && name.0 != CODEX_NAME
})
.cloned()
.collect::<Vec<_>>();
@@ -1056,8 +1056,8 @@ impl AgentConfiguration {
.gap_2()
.child(self.render_agent_server(
AgentIcon::Name(IconName::AiClaude),
- "Claude Code",
- "Claude Code",
+ "Claude Agent",
+ "Claude Agent",
ExternalAgentSource::Builtin,
cx,
))
@@ -7,12 +7,12 @@ use agent_servers::AgentServer;
use db::kvp::{Dismissable, KEY_VALUE_STORE};
use project::{
ExternalAgentServerName,
- agent_server_store::{CLAUDE_CODE_NAME, CODEX_NAME, GEMINI_NAME},
+ agent_server_store::{CLAUDE_AGENT_NAME, CODEX_NAME, GEMINI_NAME},
};
use serde::{Deserialize, Serialize};
use settings::{LanguageModelProviderSetting, LanguageModelSelection};
-use zed_actions::agent::{OpenClaudeCodeOnboardingModal, ReauthenticateAgent};
+use zed_actions::agent::{OpenClaudeAgentOnboardingModal, ReauthenticateAgent};
use crate::ManageProfiles;
use crate::ui::{AcpOnboardingModal, ClaudeCodeOnboardingModal};
@@ -195,9 +195,11 @@ pub fn init(cx: &mut App) {
.register_action(|workspace, _: &OpenAcpOnboardingModal, window, cx| {
AcpOnboardingModal::toggle(workspace, window, cx)
})
- .register_action(|workspace, _: &OpenClaudeCodeOnboardingModal, window, cx| {
- ClaudeCodeOnboardingModal::toggle(workspace, window, cx)
- })
+ .register_action(
+ |workspace, _: &OpenClaudeAgentOnboardingModal, window, cx| {
+ ClaudeCodeOnboardingModal::toggle(workspace, window, cx)
+ },
+ )
.register_action(|_workspace, _: &ResetOnboarding, window, cx| {
window.dispatch_action(workspace::RestoreBanner.boxed_clone(), cx);
window.refresh();
@@ -271,7 +273,7 @@ pub enum AgentType {
NativeAgent,
TextThread,
Gemini,
- ClaudeCode,
+ ClaudeAgent,
Codex,
Custom {
name: SharedString,
@@ -283,7 +285,7 @@ impl AgentType {
match self {
Self::NativeAgent | Self::TextThread => "Zed Agent".into(),
Self::Gemini => "Gemini CLI".into(),
- Self::ClaudeCode => "Claude Code".into(),
+ Self::ClaudeAgent => "Claude Agent".into(),
Self::Codex => "Codex".into(),
Self::Custom { name, .. } => name.into(),
}
@@ -293,7 +295,7 @@ impl AgentType {
match self {
Self::NativeAgent | Self::TextThread => None,
Self::Gemini => Some(IconName::AiGemini),
- Self::ClaudeCode => Some(IconName::AiClaude),
+ Self::ClaudeAgent => Some(IconName::AiClaude),
Self::Codex => Some(IconName::AiOpenAi),
Self::Custom { .. } => Some(IconName::Sparkle),
}
@@ -304,7 +306,7 @@ impl From<ExternalAgent> for AgentType {
fn from(value: ExternalAgent) -> Self {
match value {
ExternalAgent::Gemini => Self::Gemini,
- ExternalAgent::ClaudeCode => Self::ClaudeCode,
+ ExternalAgent::ClaudeCode => Self::ClaudeAgent,
ExternalAgent::Codex => Self::Codex,
ExternalAgent::Custom { name } => Self::Custom { name },
ExternalAgent::NativeAgent => Self::NativeAgent,
@@ -937,7 +939,7 @@ impl AgentPanel {
AgentType::NativeAgent => Some(HistoryKind::AgentThreads),
AgentType::TextThread => Some(HistoryKind::TextThreads),
AgentType::Gemini
- | AgentType::ClaudeCode
+ | AgentType::ClaudeAgent
| AgentType::Codex
| AgentType::Custom { .. } => {
if self.acp_history.read(cx).has_session_list() {
@@ -1577,7 +1579,7 @@ impl AgentPanel {
match &self.selected_agent {
AgentType::NativeAgent => Some(ExternalAgent::NativeAgent),
AgentType::Gemini => Some(ExternalAgent::Gemini),
- AgentType::ClaudeCode => Some(ExternalAgent::ClaudeCode),
+ AgentType::ClaudeAgent => Some(ExternalAgent::ClaudeCode),
AgentType::Codex => Some(ExternalAgent::Codex),
AgentType::Custom { name } => Some(ExternalAgent::Custom { name: name.clone() }),
AgentType::TextThread => None,
@@ -1644,8 +1646,8 @@ impl AgentPanel {
AgentType::Gemini => {
self.external_thread(Some(crate::ExternalAgent::Gemini), None, None, window, cx)
}
- AgentType::ClaudeCode => {
- self.selected_agent = AgentType::ClaudeCode;
+ AgentType::ClaudeAgent => {
+ self.selected_agent = AgentType::ClaudeAgent;
self.serialize(cx);
self.external_thread(
Some(crate::ExternalAgent::ClaudeCode),
@@ -2320,8 +2322,8 @@ impl AgentPanel {
.separator()
.header("External Agents")
.item(
- ContextMenuEntry::new("Claude Code")
- .when(is_agent_selected(AgentType::ClaudeCode), |this| {
+ ContextMenuEntry::new("Claude Agent")
+ .when(is_agent_selected(AgentType::ClaudeAgent), |this| {
this.action(Box::new(NewExternalAgentThread {
agent: None,
}))
@@ -2339,7 +2341,7 @@ impl AgentPanel {
{
panel.update(cx, |panel, cx| {
panel.new_agent_thread(
- AgentType::ClaudeCode,
+ AgentType::ClaudeAgent,
window,
cx,
);
@@ -2418,7 +2420,7 @@ impl AgentPanel {
.external_agents()
.filter(|name| {
name.0 != GEMINI_NAME
- && name.0 != CLAUDE_CODE_NAME
+ && name.0 != CLAUDE_AGENT_NAME
&& name.0 != CODEX_NAME
})
.cloned()
@@ -55,7 +55,10 @@ fn keywords_by_agent_feature() -> &'static BTreeMap<BuiltInAgent, Vec<&'static s
OnceLock::new();
KEYWORDS_BY_FEATURE.get_or_init(|| {
BTreeMap::from_iter([
- (BuiltInAgent::Claude, vec!["claude", "claude code"]),
+ (
+ BuiltInAgent::Claude,
+ vec!["claude", "claude code", "claude agent"],
+ ),
(BuiltInAgent::Codex, vec!["codex", "codex cli"]),
(BuiltInAgent::Gemini, vec!["gemini", "gemini cli"]),
])
@@ -325,8 +328,8 @@ impl AgentRegistryPage {
for feature in &self.upsells {
let banner = match feature {
BuiltInAgent::Claude => self.render_feature_upsell_banner(
- "Claude Code support is built-in to Zed!".into(),
- "https://zed.dev/docs/ai/external-agents#claude-code".into(),
+ "Claude Agent support is built-in to Zed!".into(),
+ "https://zed.dev/docs/ai/external-agents#claude-agent".into(),
),
BuiltInAgent::Codex => self.render_feature_upsell_banner(
"Codex CLI support is built-in to Zed!".into(),
@@ -1,6 +1,6 @@
mod acp_onboarding_modal;
mod agent_notification;
-mod claude_code_onboarding_modal;
+mod claude_agent_onboarding_modal;
mod end_trial_upsell;
mod hold_for_default;
mod mention_crease;
@@ -9,7 +9,7 @@ mod onboarding_modal;
pub use acp_onboarding_modal::*;
pub use agent_notification::*;
-pub use claude_code_onboarding_modal::*;
+pub use claude_agent_onboarding_modal::*;
pub use end_trial_upsell::*;
pub use hold_for_default::*;
pub use mention_crease::*;
@@ -8,7 +8,7 @@ use workspace::{ModalView, Workspace};
use crate::agent_panel::{AgentPanel, AgentType};
-macro_rules! claude_code_onboarding_event {
+macro_rules! claude_agent_onboarding_event {
($name:expr) => {
telemetry::event!($name, source = "ACP Claude Code Onboarding");
};
@@ -37,21 +37,21 @@ impl ClaudeCodeOnboardingModal {
if let Some(panel) = workspace.panel::<AgentPanel>(cx) {
panel.update(cx, |panel, cx| {
- panel.new_agent_thread(AgentType::ClaudeCode, window, cx);
+ panel.new_agent_thread(AgentType::ClaudeAgent, window, cx);
});
}
});
cx.emit(DismissEvent);
- claude_code_onboarding_event!("Open Panel Clicked");
+ claude_agent_onboarding_event!("Open Panel Clicked");
}
fn view_docs(&mut self, _: &ClickEvent, _: &mut Window, cx: &mut Context<Self>) {
cx.open_url(&zed_urls::external_agents_docs(cx));
cx.notify();
- claude_code_onboarding_event!("Documentation Link Clicked");
+ claude_agent_onboarding_event!("Documentation Link Clicked");
}
fn cancel(&mut self, _: &menu::Cancel, _: &mut Window, cx: &mut Context<Self>) {
@@ -172,7 +172,7 @@ impl Render for ClaudeCodeOnboardingModal {
.size(IconSize::Small)
.color(Color::Muted),
)
- .child(Label::new("New Claude Code Thread").size(LabelSize::Small)),
+ .child(Label::new("New Claude Agent Thread").size(LabelSize::Small)),
)
.child(illustration_element(
IconName::Stop,
@@ -190,11 +190,11 @@ impl Render for ClaudeCodeOnboardingModal {
.size(LabelSize::Small)
.color(Color::Muted),
)
- .child(Headline::new("Claude Code: Natively in Zed").size(HeadlineSize::Large));
+ .child(Headline::new("Claude Agent: Natively in Zed").size(HeadlineSize::Large));
- let copy = "Powered by the Agent Client Protocol, you can now run Claude Code as\na first-class citizen in Zed's agent panel.";
+ let copy = "Powered by the Agent Client Protocol, you can now run Claude Agent as\na first-class citizen in Zed's agent panel.";
- let open_panel_button = Button::new("open-panel", "Start with Claude Code")
+ let open_panel_button = Button::new("open-panel", "Start with Claude Agent")
.icon_size(IconSize::Indicator)
.style(ButtonStyle::Tinted(TintColor::Accent))
.full_width()
@@ -210,7 +210,7 @@ impl Render for ClaudeCodeOnboardingModal {
let close_button = h_flex().absolute().top_2().right_2().child(
IconButton::new("cancel", IconName::Close).on_click(cx.listener(
|_, _: &ClickEvent, _window, cx| {
- claude_code_onboarding_event!("Canceled", trigger = "X click");
+ claude_agent_onboarding_event!("Canceled", trigger = "X click");
cx.emit(DismissEvent);
},
)),
@@ -227,7 +227,7 @@ impl Render for ClaudeCodeOnboardingModal {
.overflow_hidden()
.on_action(cx.listener(Self::cancel))
.on_action(cx.listener(|_, _: &menu::Cancel, _window, cx| {
- claude_code_onboarding_event!("Canceled", trigger = "Action");
+ claude_agent_onboarding_event!("Canceled", trigger = "Action");
cx.emit(DismissEvent);
}))
.on_any_mouse_down(cx.listener(|this, _: &MouseDownEvent, window, cx| {
@@ -14,7 +14,7 @@ impl PlanDefinitions {
"Unlimited prompts with your AI API keys",
))
.child(ListBulletItem::new(
- "Unlimited use of external agents like Claude Code",
+ "Unlimited use of external agents like Claude Agent",
))
}
@@ -248,7 +248,10 @@ fn keywords_by_feature() -> &'static BTreeMap<Feature, Vec<&'static str>> {
static KEYWORDS_BY_FEATURE: OnceLock<BTreeMap<Feature, Vec<&'static str>>> = OnceLock::new();
KEYWORDS_BY_FEATURE.get_or_init(|| {
BTreeMap::from_iter([
- (Feature::AgentClaude, vec!["claude", "claude code"]),
+ (
+ Feature::AgentClaude,
+ vec!["claude", "claude code", "claude agent"],
+ ),
(Feature::AgentCodex, vec!["codex", "codex cli"]),
(Feature::AgentGemini, vec!["gemini", "gemini cli"]),
(
@@ -1567,8 +1570,8 @@ impl ExtensionsPage {
for feature in &self.upsells {
let banner = match feature {
Feature::AgentClaude => self.render_feature_upsell_banner(
- "Claude Code support is built-in to Zed!".into(),
- "https://zed.dev/docs/ai/external-agents#claude-code".into(),
+ "Claude Agent support is built-in to Zed!".into(),
+ "https://zed.dev/docs/ai/external-agents#claude-agent".into(),
false,
cx,
),
@@ -3,11 +3,6 @@ use anyhow::Result;
#[cfg(windows)]
pub fn main(_socket: &str) -> Result<()> {
// It looks like we can't get an async stdio stream on Windows from smol.
- //
- // We decided to merge this with a panic on Windows since this is only used
- // by the experimental Claude Code Agent Server.
- //
- // We're tracking this internally, and we will address it before shipping the integration.
panic!("--nc isn't yet supported on Windows");
}
@@ -460,7 +460,7 @@ impl AgentServerStore {
),
);
self.external_agents.insert(
- CLAUDE_CODE_NAME.into(),
+ CLAUDE_AGENT_NAME.into(),
ExternalAgentEntry::new(
Box::new(LocalClaudeCode {
fs: fs.clone(),
@@ -677,12 +677,12 @@ impl AgentServerStore {
// will have them.
let external_agents: [(ExternalAgentServerName, ExternalAgentEntry); 3] = [
(
- CLAUDE_CODE_NAME.into(),
+ CLAUDE_AGENT_NAME.into(),
ExternalAgentEntry::new(
Box::new(RemoteExternalAgentServer {
project_id,
upstream_client: upstream_client.clone(),
- name: CLAUDE_CODE_NAME.into(),
+ name: CLAUDE_AGENT_NAME.into(),
status_tx: None,
new_version_available_tx: None,
}) as Box<dyn ExternalAgentServer>,
@@ -909,7 +909,7 @@ impl AgentServerStore {
.map(|name| {
let agent_name = ExternalAgentServerName(name.clone().into());
let fallback_source =
- if name == GEMINI_NAME || name == CLAUDE_CODE_NAME || name == CODEX_NAME {
+ if name == GEMINI_NAME || name == CLAUDE_AGENT_NAME || name == CODEX_NAME {
ExternalAgentSource::Builtin
} else {
ExternalAgentSource::Custom
@@ -1449,26 +1449,8 @@ impl ExternalAgentServer for LocalClaudeCode {
)
.await?;
command.env = Some(env);
- let login = command
- .args
- .first()
- .and_then(|path| {
- path.strip_suffix("/@zed-industries/claude-code-acp/dist/index.js")
- })
- .map(|path_prefix| task::SpawnInTerminal {
- command: Some(command.path.to_string_lossy().into_owned()),
- args: vec![
- Path::new(path_prefix)
- .join("@anthropic-ai/claude-agent-sdk/cli.js")
- .to_string_lossy()
- .to_string(),
- "/login".into(),
- ],
- env: command.env.clone().unwrap_or_default(),
- label: "claude /login".into(),
- ..Default::default()
- });
- (command, login)
+
+ (command, None)
};
command.env.get_or_insert_default().extend(extra_env);
@@ -2229,7 +2211,7 @@ impl ExternalAgentServer for LocalCustomAgent {
}
pub const GEMINI_NAME: &'static str = "gemini";
-pub const CLAUDE_CODE_NAME: &'static str = "claude";
+pub const CLAUDE_AGENT_NAME: &'static str = "claude";
pub const CODEX_NAME: &'static str = "codex";
#[derive(Default, Clone, JsonSchema, Debug, PartialEq, RegisterSetting)]
@@ -119,7 +119,7 @@ pub struct AgentSettingsContent {
/// require confirmation.
///
/// The global `default` applies when no tool-specific rules match.
- /// For external agent servers (e.g. Claude Code) that define their own
+ /// For external agent servers (e.g. Claude Agent) that define their own
/// permission modes, "deny" and "confirm" still take precedence — the
/// external agent's permission system is only used when Zed would allow
/// the action. Per-tool regex patterns (`always_allow`, `always_deny`,
@@ -346,9 +346,9 @@ impl TitleBar {
OnboardingBanner::new(
"ACP Claude Code Onboarding",
IconName::AiClaude,
- "Claude Code",
+ "Claude Agent",
Some("Introducing:".into()),
- zed_actions::agent::OpenClaudeCodeOnboardingModal.boxed_clone(),
+ zed_actions::agent::OpenClaudeAgentOnboardingModal.boxed_clone(),
cx,
)
// When updating this to a non-AI feature release, remove this line.
@@ -434,8 +434,8 @@ pub mod agent {
OpenOnboardingModal,
/// Opens the ACP onboarding modal.
OpenAcpOnboardingModal,
- /// Opens the Claude Code onboarding modal.
- OpenClaudeCodeOnboardingModal,
+ /// Opens the Claude Agent onboarding modal.
+ OpenClaudeAgentOnboardingModal,
/// Resets the agent onboarding state.
ResetOnboarding,
/// Starts a chat conversation with the agent.
@@ -11,7 +11,7 @@ You can do that by:
1. [subscribing to our Pro plan](https://zed.dev/pricing), so you have access to our hosted models
2. [using your own API keys](./llm-providers.md#use-your-own-keys), either from model providers like Anthropic or model gateways like OpenRouter.
-3. using an [external agent](./external-agents.md) like [Gemini CLI](./external-agents.md#gemini-cli) or [Claude Code](./external-agents.md#claude-code)
+3. using an [external agent](./external-agents.md) like [Gemini CLI](./external-agents.md#gemini-cli) or [Claude Agent](./external-agents.md#claude-agent)
## Overview {#overview}
@@ -19,7 +19,8 @@ With an LLM provider or external agent configured, type in the message editor an
Responses stream in with indicators showing [which tools](./tools.md) the model is using. The sections below cover what you can do from here.
-> Note that for external agents, like [Gemini CLI](./external-agents.md#gemini-cli) or [Claude Code](./external-agents.md#claude-code), some of the features outlined below may _not_ be supported—for example, _restoring threads from history_, _checkpoints_, _token usage display_, and others. Their availability varies depending on the agent.
+> Note that for external agents, like [Gemini CLI](./external-agents.md#gemini-cli) or [Claude Agent](./external-agents.md#claude-agent), some of the features outlined below may _not_ be supported—for example, _restoring threads from history_, _checkpoints_, _token usage display_, and others.
+> Their availability varies depending on the agent.
### Creating New Threads {#new-thread}
@@ -4,8 +4,8 @@ When using AI in Zed, you can configure multiple dimensions:
1. Which LLM providers you can use
- Zed's hosted models, which require [authentication](../authentication.md) and [subscription](./subscription.md)
- - [Using your own API keys](./llm-providers.md), which do not
- - Using [external agents like Claude Code](./external-agents.md), which do not
+ - [Using your own API keys](./llm-providers.md), which do not require the above
+ - Using [external agents like Claude Agent](./external-agents.md), which also do not require the above
2. [Model parameters and usage](./agent-settings.md#model-settings)
3. [Interactions with the Agent Panel](./agent-settings.md#agent-panel-settings)
@@ -1,8 +1,13 @@
+---
+title: Use Claude Agent, Gemini CLI, and Codex in Zed
+description: Run Claude Agent, Gemini CLI, Codex, and other AI coding agents directly in Zed via the Agent Client Protocol (ACP).
+---
+
# External Agents
Zed supports many external agents, including CLI-based ones, through the [Agent Client Protocol (ACP)](https://agentclientprotocol.com).
-Zed supports [Gemini CLI](https://github.com/google-gemini/gemini-cli) (the reference ACP implementation), [Claude Code](https://www.anthropic.com/claude-code), [Codex](https://developers.openai.com/codex), [GitHub Copilot](https://github.com/github/copilot-language-server-release), and [additional agents](#add-more-agents) you can configure.
+Zed supports [Gemini CLI](https://github.com/google-gemini/gemini-cli) (the reference ACP implementation), [Claude Agent](https://platform.claude.com/docs/en/agent-sdk/overview), [Codex](https://developers.openai.com/codex), [GitHub Copilot](https://github.com/github/copilot-language-server-release), and [additional agents](#add-more-agents) you can configure.
> Note that Zed's interaction with external agents is strictly UI-based; the billing, legal, and terms arrangement is directly between you and the agent provider.
> Zed does not charge for use of external agents, and our [zero-data retention agreements/privacy guarantees](./ai-improvement.md) are **_only_** applicable for Zed's hosted models.
@@ -68,14 +73,14 @@ Gemini CLI supports the same workflows as Zed's first-party agent: code generati
> Some agent panel features are not yet available with Gemini CLI: editing past messages, resuming threads from history, and checkpointing.
-## Claude Code
+## Claude Agent
-Similar to Gemini CLI, you can also run [Claude Code](https://www.anthropic.com/claude-code) directly via Zed's [agent panel](./agent-panel.md).
-Under the hood, Zed runs Claude Code and communicate to it over ACP, through [a dedicated adapter](https://github.com/zed-industries/claude-code-acp).
+Similar to Gemini CLI, you can also run [Claude Agent](https://platform.claude.com/docs/en/agent-sdk/overview) directly via Zed's [agent panel](./agent-panel.md).
+Under the hood, Zed runs the Claude Agent SDK, which runs Claude Code under the hood, and communicates to it over ACP, through [a dedicated adapter](https://github.com/zed-industries/claude-code-acp).
### Getting Started
-Open the agent panel with {#kb agent::ToggleFocus}, and then use the `+` button in the top right to start a new Claude Code thread.
+Open the agent panel with {#kb agent::ToggleFocus}, and then use the `+` button in the top right to start a new Claude Agent thread.
If you'd like to bind this to a keyboard shortcut, you can do so by editing your `keymap.json` file via the `zed: open keymap file` command to include:
@@ -83,7 +88,7 @@ If you'd like to bind this to a keyboard shortcut, you can do so by editing your
[
{
"bindings": {
- "cmd-alt-c": ["agent::NewExternalAgentThread", { "agent": "claude_code" }]
+ "cmd-alt-c": ["agent::NewExternalAgentThread", { "agent": "claude" }]
}
}
]
@@ -91,18 +96,18 @@ If you'd like to bind this to a keyboard shortcut, you can do so by editing your
### Authentication
-As of version `0.202.7`, authentication to Zed's Claude Code installation is decoupled entirely from Zed's agent.
-That is to say, an Anthropic API key added via the [Zed Agent's settings](./llm-providers.md#anthropic) will _not_ be utilized by Claude Code for authentication and billing.
+As of version `0.202.7`, authentication to Zed's Claude Agent installation is decoupled entirely from Zed's agent.
+That is to say, an Anthropic API key added via the [Zed Agent's settings](./llm-providers.md#anthropic) will _not_ be utilized by Claude Agent for authentication and billing.
-To ensure you're using your billing method of choice, [open a new Claude Code thread](./agent-panel.md#new-thread).
+To ensure you're using your billing method of choice, [open a new Claude Agent thread](./agent-panel.md#new-thread).
Then, run `/login`, and authenticate either via API key, or via `Log in with Claude Code` to use a Claude Pro/Max subscription.
#### Installation
-The first time you create a Claude Code thread, Zed will install [@zed-industries/claude-code-acp](https://github.com/zed-industries/claude-code-acp).
+The first time you create a Claude Agent thread, Zed will install [@zed-industries/claude-code-acp](https://github.com/zed-industries/claude-code-acp).
This installation is only available to Zed and is kept up to date as you use the agent.
-Zed will always use this managed version of the Claude Code adapter, which includes a vendored version of the Claude Code CLI, even if you have it installed globally.
+Zed will always use this managed version of the Claude Agent adapter, which includes a vendored version of the Claude Code CLI, even if you have it installed globally.
If you want to override the executable used by the adapter, you can set the `CLAUDE_CODE_EXECUTABLE` environment variable in your settings to the path of your preferred executable.
@@ -120,22 +125,22 @@ If you want to override the executable used by the adapter, you can set the `CLA
### Usage
-Claude Code supports the same workflows as Zed's first-party agent. Add context by @-mentioning files, recent threads, diagnostics, or symbols.
+Claude Agent supports the same workflows as Zed's first-party agent. Add context by @-mentioning files, recent threads, diagnostics, or symbols.
-In complement to talking to it [over ACP](https://agentclientprotocol.com), Zed relies on the [Claude Code SDK](https://docs.anthropic.com/en/docs/claude-code/sdk/sdk-overview) to support some of its specific features.
+In complement to talking to it [over ACP](https://agentclientprotocol.com), Zed relies on the [Claude Agent SDK](https://platform.claude.com/docs/en/agent-sdk/overview) to support some of its specific features.
However, the SDK doesn't yet expose everything needed to fully support all of them:
- Slash Commands: A subset of [built-in commands](https://docs.anthropic.com/en/docs/claude-code/slash-commands#built-in-slash-commands) are supported, while [custom slash commands](https://docs.anthropic.com/en/docs/claude-code/slash-commands#custom-slash-commands) are fully supported.
- [Subagents](https://docs.anthropic.com/en/docs/claude-code/sub-agents) are supported.
- [Hooks](https://docs.anthropic.com/en/docs/claude-code/hooks-guide) are currently _not_ supported.
-> Some [agent panel](./agent-panel.md) features are not yet available with Claude Code: editing past messages, resuming threads from history, and checkpointing.
+> Some [agent panel](./agent-panel.md) features are not yet available with Claude Agent: editing past messages, resuming threads from history, and checkpointing.
#### CLAUDE.md
-Claude Code in Zed will automatically use any `CLAUDE.md` file found in your project root, project subdirectories, or root `.claude` directory.
+Claude Agent in Zed will automatically use any `CLAUDE.md` file found in your project root, project subdirectories, or root `.claude` directory.
-If you don't have a `CLAUDE.md` file, you can ask Claude Code to create one for you through the `init` slash command.
+If you don't have a `CLAUDE.md` file, you can ask Claude Agent to create one for you through the `init` slash command.
## Codex CLI
@@ -251,9 +256,9 @@ This lets you see the messages being sent and received between Zed and the agent
-It's helpful to attach data from this view if you're opening issues about problems with external agents like Claude Code, Codex, OpenCode, etc.
+It's helpful to attach data from this view if you're opening issues about problems with external agents like Claude Agent, Codex, OpenCode, etc.
## MCP Servers
Note that for external agents, access to MCP servers [installed from Zed](./mcp.md) may vary depending on the ACP implementation.
-For example, Claude Code and Codex both support it, but Gemini CLI does not yet.
+For example, Claude Agent and Codex both support it, but Gemini CLI does not yet.
@@ -152,5 +152,5 @@ See [Per-tool Permission Rules](./agent-settings.md#per-tool-permission-rules) a
Note that for [external agents](./external-agents.md) connected through the [Agent Client Protocol](https://agentclientprotocol.com/), access to MCP servers installed from Zed may vary depending on the ACP agent implementation.
-Regarding the built-in ones, Claude Code and Codex both support it, and Gemini CLI does not yet.
+Regarding the built-in ones, Claude Agent and Codex both support it, and Gemini CLI does not yet.
In the meantime, learn how to add MCP server support to Gemini CLI through [their documentation](https://github.com/google-gemini/gemini-cli?tab=readme-ov-file#using-mcp-servers).
@@ -2,12 +2,37 @@
Zed integrates AI throughout the editor: agentic coding, inline transformations, edit prediction, and direct model conversations.
-## Setting Up AI
+## How Zed approaches AI
-- [Configuration](./configuration.md): Connect to Anthropic, OpenAI, Ollama, Google AI, or other LLM providers.
+Zed's AI features run inside a native, GPU-accelerated application built in Rust. There is no Electron layer between you and the model output.
+
+- **Open source.** The editor and all AI features are [open source](https://github.com/zed-industries/zed). You can read how AI is implemented, how data flows to providers, and how tool calls execute.
+- **Multi-model.** Use Zed's hosted models or [bring your own API keys](./llm-providers.md) from Anthropic, OpenAI, Google, Ollama, and 8+ other providers. Run local models, connect to cloud APIs, or mix both. Switch models per task.
+- **External agents.** Run Claude Agent, Gemini CLI, Codex, and other CLI-based agents directly in Zed through the [Agent Client Protocol](https://zed.dev/acp). See [External Agents](./external-agents.md).
+- **Privacy by default.** AI data sharing is opt-in. When you use your own API keys, Zed maintains zero-data retention agreements with providers. See [Privacy and Security](./privacy-and-security.md).
+
+## Agentic editing
+
+The [Agent Panel](./agent-panel.md) is where you work with AI agents. Agents can read files, edit code, run terminal commands, search the web, and access diagnostics through [built-in tools](./tools.md).
+
+You can extend agents with additional tools through [MCP servers](./mcp.md), control what they can access with [tool permissions](./tool-permissions.md), and shape their behavior with [rules](./rules.md).
+
+The [Inline Assistant](./inline-assistant.md) works differently: select code or a terminal command, describe what you want, and the model rewrites the selection in place. It works with multiple cursors.
-- [External Agents](./external-agents.md): Run Claude Code, Codex, Aider, or other external agents inside Zed.
+## Code completions
+[Edit Prediction](./edit-prediction.md) provides AI code completions on every keystroke. Each keypress sends a request to the prediction provider, which returns single or multi-line suggestions you accept with `tab`.
+
+The default provider is Zeta, Zed's open-source model trained on open data. You can also use GitHub Copilot, Supermaven, or Codestral.
+
+## Text threads
+
+[Text Threads](./text-threads.md) are conversations with models inside any buffer. They work like a regular editor with your keybindings, multiple cursors, and standard editing features. Content is organized into message blocks with roles (You, Assistant, System).
+
+## Getting started
+
+- [Configuration](./configuration.md): Connect to Anthropic, OpenAI, Ollama, Google AI, or other LLM providers.
+- [External Agents](./external-agents.md): Run Claude Agent, Codex, Aider, or other external agents inside Zed.
- [Subscription](./subscription.md): Zed's hosted models and billing.
- [Privacy and Security](./privacy-and-security.md): How Zed handles data when using AI features.
@@ -311,7 +311,7 @@ To use other AI models in Zed, you have several options:
- Use Zed's hosted models, with higher rate limits. Requires [authentication](https://zed.dev/docs/accounts.html) and subscription to [Zed Pro](https://zed.dev/docs/ai/subscription.html).
- Bring your own [API keys](https://zed.dev/docs/ai/llm-providers.html), no authentication needed
-- Use [external agents like Claude Code](https://zed.dev/docs/ai/external-agents.html)
+- Use [external agents like Claude Agent](https://zed.dev/docs/ai/external-agents.html)
## Advanced Config and Productivity Tweaks
@@ -379,7 +379,7 @@ To use other AI models in Zed, you have several options:
- Use Zed's hosted models, with higher rate limits. Requires [authentication](https://zed.dev/docs/accounts.html) and subscription to [Zed Pro](https://zed.dev/docs/ai/subscription.html).
- Bring your own [API keys](https://zed.dev/docs/ai/llm-providers.html), no authentication needed
-- Use [external agents like Claude Code](https://zed.dev/docs/ai/external-agents.html)
+- Use [external agents like Claude Agent](https://zed.dev/docs/ai/external-agents.html)
## Advanced Config and Productivity Tweaks
@@ -402,7 +402,7 @@ To use other AI models in Zed, you have several options:
- Use Zed's hosted models, with higher rate limits. Requires [authentication](https://zed.dev/docs/accounts.html) and subscription to [Zed Pro](https://zed.dev/docs/ai/subscription.html).
- Bring your own [API keys](https://zed.dev/docs/ai/llm-providers.html), no authentication needed
-- Use [external agents like Claude Code](https://zed.dev/docs/ai/external-agents.html)
+- Use [external agents like Claude Agent](https://zed.dev/docs/ai/external-agents.html)
## Advanced Config and Productivity Tweaks
@@ -338,7 +338,7 @@ To use other AI models in Zed, you have several options:
- Use Zed’s hosted models, with higher rate limits. Requires [authentication](https://zed.dev/docs/accounts.html) and subscription to [Zed Pro](https://zed.dev/docs/ai/subscription.html).
- Bring your own [API keys](https://zed.dev/docs/ai/llm-providers.html), no authentication needed
-- Use [external agents like Claude Code](https://zed.dev/docs/ai/external-agents.html).
+- Use [external agents like Claude Agent](https://zed.dev/docs/ai/external-agents.html).
### Advanced Config and Productivity Tweaks
@@ -380,7 +380,7 @@ To use other AI models in Zed, you have several options:
- Use Zed's hosted models, with higher rate limits. Requires [authentication](https://zed.dev/docs/accounts.html) and subscription to [Zed Pro](https://zed.dev/docs/ai/subscription.html).
- Bring your own [API keys](https://zed.dev/docs/ai/llm-providers.html), no authentication needed
-- Use [external agents like Claude Code](https://zed.dev/docs/ai/external-agents.html)
+- Use [external agents like Claude Agent](https://zed.dev/docs/ai/external-agents.html)
## Advanced Config and Productivity Tweaks