When `use_modifier_to_send` is turned on, holding `cmd`/`ctrl` is
necessary to send a message in the agent panel. Text threads already use
`cmd-enter` by default to submit a message, and it was done this way to
have the usual text editing bindings not taken over when writing a
prompt, sort of stimulating more thoughtful writing. While `enter` to
send is still somewhat a huge pattern in chat-like LLM UIs, it still
makes sense to allow this for the new agent panel... hence the existence
of this setting now!
Release Notes:
- agent: Added the `use_modifier_to_send` setting, which makes holding a
modifier (`cmd`/`ctrl`), together with `enter`, required to send a new
message.
This includes making sure that both the agent panel and Zed's edit
prediction have a consistent narrative when it comes to onboarding users
into the AI features, considering the possible different plans and
conditions (such as being signed in/out, account age, etc.)
Release Notes:
- N/A
---------
Co-authored-by: Bennet Bo Fenner <53836821+bennetbo@users.noreply.github.com>
Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>
Release Notes:
- agent: Fixed an issue where clicking on "Start New Thread" in the
agent configuration would not always switch to the correct
provider/model
Now we only auto-retry if burn mode is enabled. We also show a "Retry"
button (so you don't have to type "continue") if you think that's the
right remedy, and additionally we show a "Retry and Enable Burn Mode"
button if you don't have it enabled.
<img width="484" height="260" alt="Screenshot 2025-07-17 at 6 25 27 PM"
src="https://github.com/user-attachments/assets/dc5bf1f6-8b11-4041-87aa-4f37c95ea9f0"
/>
<img width="478" height="307" alt="Screenshot 2025-07-17 at 6 22 36 PM"
src="https://github.com/user-attachments/assets/1ed6578a-1696-449d-96d1-e447d11959fa"
/>
Release Notes:
- Only auto-retry Agent requests when Burn Mode is enabled
Closes#34319
In this pr: https://github.com/zed-industries/zed/pull/33462 there was
check added for early return for active_thread and message_editor as
those are not present in the TextThread and only available in the Thread
the token count was not getting triggered for TextThread this pr fixes
that regression by moving the logic specific to Thread inside of thread
view match.
<img width="3024" height="1886" alt="CleanShot 2025-07-15 at 23 50
18@2x"
src="https://github.com/user-attachments/assets/bd74ae8b-6c37-4cdd-ab95-d3c253b8a948"
/>
Release Notes:
- Fix token count not getting shown in the TextThread
In #33275 I was very conservative about when to retry when there are
errors in language completions in the Agent panel.
Now we retry in more scenarios (e.g. HTTP 5xx and 4xx errors that aren't
in the specific list of ones that we handle differently, such as 429s),
and also we show a notification if the thread halts for any reason.
<img width="441" height="68" alt="Screenshot 2025-07-15 at 12 51 30 PM"
src="https://github.com/user-attachments/assets/433775d0-a8b2-403d-9427-1e296d164980"
/>
<img width="482" height="322" alt="Screenshot 2025-07-15 at 12 44 15 PM"
src="https://github.com/user-attachments/assets/5a508224-0fe0-4d34-9768-25d95873eab8"
/>
Release Notes:
- Automatic retry for more Agent errors
- Whenever the Agent stops, play a sound (if configured) and show a
notification (if configured) if the Zed window was in the background.
This PR aims to improve the minimap performace. This is primarily
achieved by disabling/removing stuff that is not shown in the minimal as
well as by assuring the display map is not updated during minimap
prepaint.
This should already be much better in parts, as the block map as well as
the fold map will be less frequently updated due to the minimap
prepainting (optimally, they should never be, but I think we're not
quite there yet).
For this, I had to remove block rendering support for the minimap, which
is not as bad as it sounds: Practically, we were currently not rendering
most blocks anyway, there were issues due to this (e.g. scrolling any
visible block offscreen in the main editor causes scroll jumps
currently) and in the long run, the minimap will most likely need its
own block map or a different approach anyway. The existing
implementation caused resizes to occur very frequently for practically
no benefit. Can pull this out into a separate PR if requested, most
likely makes the other changes here easier to discuss.
This is WIP as we are still hitting some code path here we definitely
should not be hitting. E.g. there seems to be a rerender roughly every
second if the window is unfocused but visible which does not happen when
the minimap is disabled.
While this primarily focuses on the minimap, it also touches a few other
small parts not related to the minimap where I noticed we were doing too
much stuff during prepaint. Happy for any feedback there aswell.
Putting this up here already so we have a place to discuss the changes
early if needed.
Release Notes:
- Improved performance with the minimap enabled.
- Fixed an issue where interacting with blocks in the editor would
sometimes not properly work with the minimap enabled.
When the panel was zoomed in, the edit bar's background color would have
a different color than the rest of the panel. This PR fixes it by using
the `panel_background` color token.
Release Notes:
- N/A
This introduces a new field `thinking_allowed` on `LanguageModelRequest`
which lets us control whether thinking should be enabled if the model
supports it.
We permit thinking in the Inline Assistant, Edit File tool and the Git
Commit message generator, this should make generation faster when using
a thinking model, e.g. `claude-sonnet-4-thinking`
Release Notes:
- N/A
Just a small quality-of-life type of PR that makes clicking anywhere
until the "Review" button trigger the action that that button triggers
(i.e., opens the review multibuffer).
<img
src="https://github.com/user-attachments/assets/8936ed75-50fd-49f9-89ef-b6c4301a8eba"
width="600" />
Release Notes:
- agent: Added the ability to click the whole file row in the edits bar
to trigger the review multibuffer.
Previously if I set enabled: false for one the context servers in
settings.json it will not show up in the settings in agent panel when I
start zed. But if I enabled it from settings it properly showed up. We
were filtering the configuration to only get the enabled context servers
from settings.json. This PR adds fetching all of them.
Release Notes:
- agent: Show context servers which are disabled in settings in agent
panel settings.
So we can use it in other places and don't require them to depend on the
`agent_ui`. This PR also renames it from `AnimatedLabel` to
`LoadingLabel`.
Release Notes:
- N/A
Some design polish here as a follow-up to making the provider accordion
header entirely clickable.
Release Notes:
- agent: Improved design in the provider section by refining spacing.
Closes #ISSUE
Adds a new `documentation` method to actions, that is extracted from doc
comments when using the `actions!` or derive macros.
Additionally, this PR adds doc comments to as many action definitions in
Zed as possible.
Release Notes:
- N/A *or* Added/Fixed/Improved ...
Closes#33792
Follow up to #33237 - Turns out my fix for this was not correct
Release Notes:
- agent: Fixed an issue where tools would not work when two MCP servers
provided a tool with the same name
This PR makes all errors in the agent panel to use the `Callout`
component instead of toasts. Reason for that is because the toasts
obscured part of the panel's UI, which wasn't ideal. We can also be more
expressive here with a background color, which I think helps with
parsing the message.
Release Notes:
- agent: Improved how we display errors in the panel.
As we are in the process of improving our Onboarding UX for Zed AI, I
added component previews for the Zed AI Configuration section. This
should make it easier to inspect the different states we can run into.
<img width="1198" alt="image"
src="https://github.com/user-attachments/assets/eb774f27-9091-450d-bfae-c688d533c25e"
/>
Release Notes:
- N/A
* Updates to `zed_llm_client-0.8.5` which adds support for `retry_after`
when anthropic provides it.
* Distinguishes upstream provider errors and rate limits from errors
that originate from zed's servers
* Moves `LanguageModelCompletionError::BadInputJson` to
`LanguageModelCompletionEvent::ToolUseJsonParseError`. While arguably
this is an error case, the logic in thread is cleaner with this move.
There is also precedent for inclusion of errors in the event type -
`CompletionRequestStatus::Failed` is how cloud errors arrive.
* Updates `PROVIDER_ID` / `PROVIDER_NAME` constants to use proper types
instead of `&str`, since they can be constructed in a const fashion.
* Removes use of `CLIENT_SUPPORTS_EXA_WEB_SEARCH_PROVIDER_HEADER_NAME`
as the server no longer reads this header and just defaults to that
behavior.
Release notes for this is covered by #33275
Release Notes:
- N/A
---------
Co-authored-by: Richard Feldman <oss@rtfeldman.com>
Co-authored-by: Richard <richard@zed.dev>
In #32656 I generalized the argument to change selections to allow
controling both the scroll and the nav history (and the completion
trigger).
To avoid conflicting with ongoing debugger cherry-picks I left the
argument as an `impl Into<>`, but I think it's clearer to make callers
specify what they want here.
I converted a lot of `None` arguments to `SelectionEffects::no_scroll()`
to be exactly compatible; but I think many people used none as an "i
don't care" value in which case Default::default() might be more
appropraite
Closes #ISSUE
Release Notes:
- N/A
