Addressing the feedback left on a previous PR of mine:
https://github.com/zed-industries/zed/pull/46663. I feel like this is
worthy of a setting. Note, however, that this does not apply to typing
`ctrl-c` in the terminal from within the agent panel; this is purely
controlling what happens when you click on the "Stop" button in the
terminal tool call card.
Release Notes:
- Agent: Added a setting for controlling the behavior of the stop button
in the terminal tool call card (between cancelling the command to run
and the thread, or just the command).
<img width="677" height="218" alt="Screenshot 2026-01-23 at 11 59 24 AM"
src="https://github.com/user-attachments/assets/864c706f-6f9f-44dd-a18c-509f2bed164f"
/>
When a thread is interrupted while subagents are running, the subagent
cards now show a red X icon instead of a green checkmark. This provides
clearer visual feedback that the subagent was canceled rather than
completed successfully.
The icon logic now handles three states:
- **Spinner**: when status is Pending or InProgress
- **Red X**: when status is Canceled, Failed, or Rejected
- **Green checkmark**: when status is Completed
This matches the existing pattern used elsewhere in the codebase for
showing error states on tool calls.
(No release notes because subagents are still behind a feature flag.)
Release Notes:
- N/A
- **title_bar: Extract platform_title_bar from title_bar**
- **file_finder no longer depends on search and agent_servers no longer
depend on language_models**
Release Notes:
- N/A
---------
Co-authored-by: Zed Zippy <234243425+zed-zippy[bot]@users.noreply.github.com>
The rate limiter's semaphore guard was being held for the entire
duration of a turn, including during tool execution. This caused
deadlocks when subagents tried to acquire permits while parent requests
were waiting for them to complete.
## The Problem
In `run_turn_internal`, the stream (which contains the `RateLimitGuard`
holding the semaphore permit) was kept alive throughout the entire loop
iteration - including during **tool execution**:
1. Parent request acquires permit
2. Parent starts streaming, consumes response
3. Parent starts executing tools (subagents)
4. **Stream/guard still held** while tools execute
5. Subagents try to acquire permits → blocked because parent still holds
permit
6. Deadlock if all permits are held by parents waiting for subagent
children
## The Fix
Two changes were made:
1. **Drop the stream early**: Added an explicit `drop(events)` after the
stream is fully consumed but before tool execution begins. This releases
the rate limit permit so subagents can acquire it.
2. **Removed the `bypass_rate_limit` workaround**: Since the root cause
is now fixed, the bypass mechanism is no longer needed.
Note: no release notes because subagents are still feature-flagged, and
this rate limiting change isn't actually observable without them.
Release Notes:
- N/A
This is behind the beta flag. But some agents don't support a full load
flow, but rather just a resume from the current state of the session.
They don't emit notifications for past thread messages, which isn't
ideal, but the user can at least resume their work without completely
losing it.
We use it as a fallback if the agent doesn't support loading only, and
signal to the user that they won't be able to see previous messages.
Release Notes:
- N/A
Closes https://github.com/zed-industries/zed/issues/47330
This PR moves the queue logic out of the native Thread into the shared
UI layer (thread view) to enable it for external agents. There's a key
difference in behavior, though, between native and external agents:
queued messages in the former will be sent in the next turn boundary,
given we can easily tell this, whereas for the latter, queued messages
will be sent by the end of the generation. We'd need an ACP-level change
to provide exactly the same UX between both types of agents, and I
figured that's better to have _some_ version of the feature for external
agents as opposed to not having it all due to this difference.
Release Notes:
- Agent: Made the message queue feature available for external agents as
well.
This PR adds a thinking toggle for controlling whether to use thinking
for a model in the Zed provider:
<img width="645" height="142" alt="Screenshot 2026-01-22 at 12 34 01 PM"
src="https://github.com/user-attachments/assets/9aa543fe-e708-4840-8b38-1a6fbcb78388"
/>
Previously we would create separate "Thinking" variants of the models
that supported thinking in the model selector.
This only applies to Anthropic models in the Zed provider, currently.
This is gated behind the `cloud-thinking-toggle` feature flag.
Release Notes:
- N/A
---------
Co-authored-by: Neel <neel@zed.dev>
Adds a new "Agent Panel Error Shown" telemetry event that fires when
users see errors in the agent panel. For errors from external ACP agents
(like Claude Code), we capture additional details.
Previously, we had no visibility into what errors users were
encountering in the agent panel. This made it difficult to diagnose
issues, especially with external agents.
The new telemetry event includes:
- `agent` — The agent telemetry ID
- `session_id` — The session ID
- `kind` — Error category (payment_required, authentication_required,
refusal, other, etc.)
- `acp_error_code` — The ACP error code when available (e.g.,
"InternalError")
- `acp_error_message` — The ACP error message when available
Release Notes:
- N/A
---------
Co-authored-by: Michael Benfield <mbenfield@zed.dev>
Closes https://github.com/zed-industries/zed/issues/47347
Release Notes:
- Agent: Fixed a bug where the inline assistant feedback buttons where
being displayed despite the setting "Enable Feedback" was turned off.
Some users were a bit confused that the history between zed agent + text
threads is split up now. Making it a bit clearer for the users still
using text threads which history they are currently looking at.
Release Notes:
- N/A
Closes https://github.com/zed-industries/zed/issues/41278,
https://github.com/zed-industries/zed/issues/45576,
https://github.com/zed-industries/zed/issues/46513
Release Notes:
- Fixed message editor stealing focus (so the "Open Recent Project"
modal stays open)
This issue has been driving me insane haha.
_Note: This is my first code contribution to this project. Please be
kind and let me know if I have missed something. I tested this fix on my
machine and it works. Tests are passing on my device. I have agreed to
the Zed Contributor License and Feedback Agreement with my Zed Account
that's linked to the same email as this GitHub account._
This PR implements a UI for the side-by-side diff, using blocks to align
the two sides and adding a coherent `SplitEditorElement`.
Release Notes:
- N/A
---------
Co-authored-by: cameron <cameron.studdstreet@gmail.com>
Co-authored-by: Anthony Eid <hello@anthonyeid.me>
Co-authored-by: Zed Zippy <234243425+zed-zippy[bot]@users.noreply.github.com>
## Summary
Implements user-defined slash commands for the agent panel. Users can
create markdown files in `~/.config/zed/commands/` (user-wide) or
`.zed/commands/` (project-specific) that expand into templated text when
invoked via `/command_name` in the chat interface.
## Features
- **File-based commands**: Create `.md` files that become slash commands
- **Template expansion**: Use `$1`, `$2`, etc. for positional arguments,
or `$ARGUMENTS` for all args
- **Namespacing**: Subdirectories create namespaced commands (e.g.,
`frontend/component.md` → `/frontend:component`)
- **Project & user scopes**: Project commands in `.zed/commands/`, user
commands in config dir
- **Claude compatibility**: Can symlink `~/.claude/commands/` for
compatibility
- **Caching**: Commands are cached and watched for file changes
- **Error handling**: Graceful degradation with error display in UI
## Feature Flag
Behind `user-slash-commands` feature flag.
## Testing
52 unit/integration tests covering parsing, validation, expansion, file
loading, symlinks, and error handling.
---
Release Notes:
- N/A (behind feature flag)
---------
Co-authored-by: Zed Zippy <234243425+zed-zippy[bot]@users.noreply.github.com>
This PR improves the performance of the `CopyButton` component by
tracking the copied state locally through a `CopyButtonState` struct
instead of making an OS call every time the component re-renders. Also
pushing a slight improvement here by resetting the state after two
seconds so as to make the check mark go away after you clicked to copy.
Release Notes:
- N/A
- Improves spacing between consecutive calls
- Only show disclosure button (i.e., ability to expand the card) when
there's actual content
- Fix font size being bigger than the actual thread
- Add gradient overlay to communicate that you can't scroll within the
card (expanding the subagent call will allow that in the near future)
Here's the result of it all, roughly:
<img width="600" height="890" alt="Screenshot 2026-01-21 at 12 26@2x"
src="https://github.com/user-attachments/assets/e5e4a544-611d-48f5-b00d-f69e869726c5"
/>
Release Notes:
- N/A
## Problem
When subagents use the `edit_file` tool, it creates an `EditAgent` that
makes its own model request to get the edit instructions. These "nested"
requests compete with the parent subagent conversation requests for rate
limiter permits.
The rate limiter uses a semaphore with a limit of 4 concurrent requests
per model instance. When multiple subagents run in parallel:
1. 3 subagents each hold 1 permit for their ongoing conversation streams
(3 permits used)
2. When all 3 try to use `edit_file` simultaneously, their edit agents
need permits too
3. Only 1 edit agent can get the 4th permit; the other 2 block waiting
4. The blocked edit agents can't complete, so their parent subagent
conversations can't complete
5. The parent conversations hold their permits, so the blocked edit
agents stay blocked
6. **Deadlock**
## Solution
Added a `bypass_rate_limit` field to `LanguageModelRequest`. When set to
`true`, the request skips the rate limiter semaphore entirely. The
`EditAgent` sets this flag because its requests are already "part of" a
rate-limited parent request.
(No release notes because subagents are still feature-flagged.)
Release Notes:
- N/A
---------
Co-authored-by: Zed Zippy <234243425+zed-zippy[bot]@users.noreply.github.com>
This PR adds the ability to edit a queued message, which you can now do
by hitting `cmd-e` from the message editor, which will focus the first
queued message. To pull that off, I'm also making the queued messages
render as an editor, the same way we do with regular user messages. That
way, we ensure less layout shift when focusing in and out of the queued
message for editing and gain the ability to render context
buttons/creases the same way we do in the main message editor.
https://github.com/user-attachments/assets/fb68fd48-c0cd-491f-a7d9-5065a9151b0b
Note that in the video, I show the state in which you're still editing
in the moment in which the queued message would be sent. If that
happens, your queued message won't be sent even if you unfocus the
queued message editor. In this case, you need to explicitly hit "Send
Now".
Release Notes:
- Agent: Added the ability to edit queued messages.
This is still behind a feature flag as the registry is still WIP, but
allows downloading binary agents from the registry on github.
Release Notes:
- N/A
Closes#31351
# Diagnostics Mention in New Threads
## Overview
Adds the `@diagnostics` mention to the new Agent Panel threads so users
can inject current LSP diagnostics (errors by default) without switching
to a text thread. The diagnostics mention is fully integrated into ACP
mention parsing, the context picker, and the message-editor pipeline so
it round-trips cleanly and shows up in the standard `@` menu.
## Context
- **Request:** bring `/diagnostics` parity to the “New Thread” assistant
experience.
- **Scope:** diagnostics only; `/terminal` mention would be implemented
in a separate PR.
- **Docs:** updated Agent Panel docs + changelog.
## Implementation Details
1. **Mention plumbing**
- Added `MentionUri::Diagnostics` to `acp_thread`, including parsing
(`zed:///agent/diagnostics?include_warnings=true`) and icon/name
metadata.
- Tests ensure diagnostics links round-trip via Markdown mention
serialization.
2. **Context picker / completion**
- New `ContextPickerMode::Diagnostics` exposes an `@diagnostics` entry
in the mention menu.
- Completions turn `@diagnostics` into a fully fledged mention, reusing
the existing confirmation pipeline.
3. **Message editor + thread serialization**
- Resolving the mention calls the existing diagnostics collector from
`assistant_slash_commands`, embedding the tool output inline with other
context blocks (`<diagnostics>…</diagnostics>`).
- Thread-link handling ignores diagnostics backlinks so clicking them
doesn’t try to reopen nonexistent resources.
# How it looks
<img width="800" height="480" alt="image"
src="https://cf5gpe8lxo.ufs.sh/f/EmJ5Xl877qJO1mzC9Zrn8AmJZHeShC4RoUwvTMlF2tfPzj06"
/>
Release Notes:
- Allow mentioning diagnostics in the agent panel via `@diagnostics`
---------
Co-authored-by: Bennet Bo Fenner <bennet@zed.dev>
Follow up to https://github.com/zed-industries/zed/pull/40570.
This PR removes the UI treatment where we collapse a terminal command
into a disclosure component if it spans more than three lines. I
initially thought having this fancy treatment in the UI would make it
sleeker but it actually feels like a counter-productive move to
arbitrarily hide command lines; they're arguably always important to see
in full, even when they're super long. I'm happy to get some feedback
here, but I think for now, I'd feel better if users could see them all
without having to manually expand/collapse them.
Release Notes:
- N/A
This makes sure all of the new granular permission logic and ui only
applies to the zed agent and doesn't affect the UI of external agents.
Release Notes:
- N/A
This feature cost $12
Release Notes:
- Changed the behavior of queued messages to gracefully wait for the
current activity to complete.
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
This PR fixes the fact that we weren't showing any @-mentions (files,
images, etc.) in the queue preview.
Release Notes:
- Agent: FIxed a bug with the message queue in the agent panel where
@-mentions weren't being displayed in the message preview within the
queue entries list.
Closes #ISSUE
Moves the settings content definitions into their own crate, so that
they are compiled+cached separately from settings, primarily to avoid
recompiles due to changes in gpui. In that vain many gpui types such as
font weight/features, and `SharedString` were replaced in the content
crate, either with `*Content` types for font/modifier things, or
`String`/`Arc<str>` for `SharedString`. To make the conversions easy a
new trait method in the settings crate named `IntoGpui::into_gpui`
allows for `into()` like conversions to the gpui types in
`from_settings` impls.
Release Notes:
- N/A
---------
Co-authored-by: Piotr Osiewicz <24362066+osiewicz@users.noreply.github.com>
This PR removes the code for the legacy plans.
No more users will be on this plan as of January 17th, so it's fine to
land these changes now (as they won't be released until the 21st).
Closes CLO-76.
Release Notes:
- N/A
Happy New Years! This PR is a second take at
https://github.com/zed-industries/zed/pull/38127 (cc @ConradIrwin)
This PR is significantly less complicated than the last attempt: while
we still keep our data on the `NavigationHistory` object, we no longer
tightly integrate it with the existing back/forward "browser history."
Instead, we keep our own stack of `(origin, target)` pairs (in a struct
to make it easy to extend with e.g., tag names in the future).
The PR is split into two separable commits. Most of the implementation
is in the second commit, which:
- Defines the stack data structure
- Implements `pane::GoToOlderTag` and `pane::GoToNewerTag` in terms of
the stack
- Hooks into `navigate_to_hover_links` to push tag stack entries
This last bit is the most fiddly. The core challenge is that we need to
keep track of the `origin` location and calculate the `target` location
across three codepaths that might involve creating a new editor and/or
splitting the pane. One thing in particular I found difficult was that
an editor's `nav_history` (an `ItemNavHistory`) seems to be populated
asynchronously. Instead of relying on it, I decided in this code to make
my own `ItemNavHistory`. I briefly tried to refactor the code in
question, but it seemed like it would significantly increase the scope
of the change.
I prefer this all-in-one tracking centered around
`navigate_to_hover_links ` to the `start/finish` approach taken in
https://github.com/zed-industries/zed/commit/b69a2ea200561de9de0edf4d0c4428ee887fb918
because I find it easier to convince myself that the right data is being
populated at the right times. Of course, let me know if you think
there's a better solution.
Closes#14206
Release Notes:
- ??? I don't know what to write here! Suggestions welcome
Release Notes:
- Opening bundled files, keymap, and local release notes now opens in
remote windows instead of opening a new local zed window
- Opening the settings files, keymap files, task files, debug files and
logs will now open within wsl windows instead of opening a new local zed
window
This feature cost $15.
Up -> Tokens we're sending to the model
Down -> Tokens we've received from the model.
<img width="377" height="69" alt="Screenshot 2026-01-14 at 12 31 01 PM"
src="https://github.com/user-attachments/assets/fc15824f-de5d-466b-8cc1-329f3c1940bb"
/>
Release Notes:
- Changed the display of tokens for OpenAI models to reflect the
input/output limits.
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Ben Brandt