This reverts commit f12a554f86, which
introduced a panic in inline assistant (cc @mgsloan) - I'm not sure what
the motivation was for that change, but I figure we can revert to fix
the inline assistant now and deal with that later. 😄
Panic was:
> Thread "main" panicked with "cannot read workspace::Workspace while it
is already being updated" at
/Users/rtfeldman/code/zed/crates/gpui/src/app/entity_map.rs:139:32
Release Notes:
- N/A
Trying out sending the model a reminder message about code blocks in the
system prompt. If this seems to work well, we can include more specific
reminder messages, e.g. tool-specific ones.
Release Notes:
- N/A
Sometimes agents do this. I've had some success responding by telling it
not to do this, so trying out having it in the system prompt.
Release Notes:
- Adjusted the system prompt to avoid incomplete code generation.
Currently, it's pretty common that when the agent gets stuck, it deletes
whatever it's stuck on and replaces it with a TODO comment, then
cheerfully reports that it has "simpified" the implementation. This is
worse than leaving the broken code, because at least a human could take
over and try to get it across the finish line.
This system prompt adjustment attempts to make the agent do something
more useful when in this situation: report that it's stuck, explain why
it's stuck, and ask the user what to do.
Release Notes:
- N/A
<img width="622" alt="Screenshot 2025-04-05 at 5 48 14 PM"
src="https://github.com/user-attachments/assets/24b9c7d4-d3e2-4929-bca8-79db5b4e5748"
/>
Release Notes:
- The `read_files` tool now reads only the symbol outline files above a
certain size, to conserve context window space. Then it suggests that
the agent call `read_files` again with the relevant line ranges it saw
in the outline.
Lets you get all the code symbols in the project (like the Code Symbols
panel) or in a particular file (like the Outline panel), optionally
paginated and filtering results by regex. The tool gives the files,
lines, and numbers of all of these, which means they can be used in
conjunction with the read file tool to read subsets of large files
without having to open the entire large file and poke around in it.
<img width="621" alt="Screenshot 2025-03-29 at 12 00 21 PM"
src="https://github.com/user-attachments/assets/d78259d7-2746-44c0-ac18-2e21f2505c0a"
/>
Release Notes:
- N/A
By default, agent notifications now display only on your primary screen.
You can optionally configure them to display on all screens (or not to
display at all).
Release Notes:
- N/A
This caused undesirable and unnecessary scrolling, in that if you scroll
up, then as new messages stream into the panel, they jump the scroll bar
back down.
@agu-z and I paired on this and we think this is unnecessary now that we
don't see the Edit button in the UI, but @bennetbo we still see code for
the button in there, so...maybe we do still want this? (If so, we can
revert the second commit and go back to a more conditional way of
scrolling. 😄)
Release Notes:
- N/A
---------
Co-authored-by: Agus <agus@zed.dev>
<img width="620" alt="Screenshot 2025-03-27 at 2 29 13 PM"
src="https://github.com/user-attachments/assets/dd023507-61bc-4722-a095-f65f4b6c746a"
/>
We'll iterate on the UI, but first the goal is to just get it to work at
all so we can see if it's useful in terms of getting correct output
faster.
Release Notes:
- N/A
---------
Co-authored-by: Agus Zubiaga <hi@aguz.me>
Now if you refocus Zed manually (e.g. cmd-tab), we hide the "View Panel"
notification automatically.
This also fixes a related subscription leak.
Release Notes:
- N/A
@agu-z and paired on trying out a "one tool call per edit" approach for
editing files. (The previous approach is still available, it's just
unchecked by default for now.)
Release Notes:
- N/A
---------
Co-authored-by: Agus <agus@zed.dev>
Sometimes we've seen models provide an empty string for the path search
glob. This assumes they meant "*" when that happens.
Separately, this also removes an unnecessary `clone` of a `String`.
Release Notes:
- N/A
We decided to take this out for now. It doesn't seem necessary, and it
complicates the code a lot. We can always put it back later if desired.
Release Notes:
- N/A
If the tool asks to read a path, we don't need to verify whether that
path exists on disk; an unsaved buffer with that path is fine.
Release Notes:
- N/A
Per a conversation with @nathansobo, have the Lua scripts run
unsandboxed for now (while this feature is behind the staff feature
flag).
Release Notes:
- N/A
Just a basic implementation so we can start trying it out.
Release Notes:
- N/A
---------
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Michael <michael@zed.dev>
Note that this does *not* involve any breaking code changes.
cc @0xtimsb - I didn't change any settings or anything here. That can
happen separately!
Release Notes:
- N/A
> **Note:** https://github.com/zed-industries/zed/pull/23813 should be
merged first!
@nathansobo and I paired on revising this menu, including adding the
"Predict Edits at Cursor" menu item (to make the keyboard shortcut more
discoverable; clicking it makes the inline edits show up, as shown in
the second screenshot) and switching from "Hide/Show" language to
checkboxes.
## Before
<img width="282" alt="Screenshot 2025-01-28 at 4 51 37 PM"
src="https://github.com/user-attachments/assets/309c82c1-8fb5-44db-950e-1a8789a63993"
/>
## After
<img width="1138" alt="Screenshot 2025-01-28 at 4 50 05 PM"
src="https://github.com/user-attachments/assets/302a126c-9389-42a4-bb7d-2896bce859e7"
/>
We also switched to use `SharedString` in more places, where it made
more sense.
@danilo-leal This isn't necessarily *exactly* what we want, but we were
pairing and decided to get it in a state where we can actually try it
out and tweak from here.
Release Notes:
- N/A
---------
Co-authored-by: Nathan <nathan@zed.dev>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Marshall Bowers <git@maxdeviant.com>
`@mention`ed files in assistant2 now get replaced by the full path of
the file in what gets sent to the model, while rendering visually as
just the filename (in a crease, so they can only be selected/deleted as
a whole unit, not character by character).
https://github.com/user-attachments/assets/a5867a93-d656-4a17-aced-58424c6e8cf6
Release Notes:
- N/A
---------
Co-authored-by: João Marcos <joao@zed.dev>
Co-authored-by: Conrad <conrad@zed.dev>
Before, the conversation popover menu covered up what you were typing
because it wasn't offset properly.
Now it's offset properly, using the UI font size so the amount of offset
scales with the font size:
<img width="435" alt="Screenshot 2025-01-07 at 4 34 27 PM"
src="https://github.com/user-attachments/assets/55e40910-8cd4-4548-b4fb-521eb2845775"
/>
<img width="454" alt="Screenshot 2025-01-07 at 4 33 58 PM"
src="https://github.com/user-attachments/assets/30350489-09f1-4cb8-9f95-ed4ee87bc110"
/>
<img width="488" alt="Screenshot 2025-01-07 at 4 34 18 PM"
src="https://github.com/user-attachments/assets/de60d990-2bd9-418d-a616-56beb3e4aa8a"
/>
Release Notes:
- N/A *or* Added/Fixed/Improved ...
Also makes the inline assistant and inline terminal assistant share a
bunch more code.
Release Notes:
- N/A
---------
Co-authored-by: Agus <agus@zed.dev>
Co-authored-by: Agus Zubiaga <hi@aguz.me>
This is behind the Assistant v2 feature flag. As @maxdeviant and I
discussed, the state is currently decoupled from the Assistant Panel's
state, although in the future we plan to introduce a way to refer to
conversations from the panel. Also, we're intentionally duplicating some
code with the v2 panel right now; the plan is to do a future PR to make
them share code more.
https://github.com/user-attachments/assets/bb163bd3-a02d-4a91-8f8f-2a8e60acbc34
It doesn't include the terminal inline assistant, which will be in a
separate PR.
Release Notes:
- N/A
Change the "Apply" and "Discard" buttons to match @danilo-leal's design!
Here are some different states:
### Cursor in the first hunk
Now that the cursor is in a particular hunk, we show the "Apply" and
"Discard" names, and the keyboard shortcut. If I press the keyboard
shortcut, it will only apply to this hunk.
<img width="759" alt="Screenshot 2024-11-23 at 10 54 45 PM"
src="https://github.com/user-attachments/assets/68e0f109-9493-4ca2-a99c-dfcbb4d1ce0c">
### Cursor in the second hunk
Moving the cursor to a different hunk changes which buttons get the
keyboard shortcut treatment. Now the keyboard shortcut is shown next to
the hunk that will actually be affected if you press that shortcut.
<img width="749" alt="Screenshot 2024-11-23 at 10 56 27 PM"
src="https://github.com/user-attachments/assets/59c2ace3-6972-4a60-b806-f45e8c25eaae">
Release Notes:
- Restyled Apply/Discard buttons
---------
Co-authored-by: Max <max@zed.dev>
Co-authored-by: Danilo Leal <67129314+danilo-leal@users.noreply.github.com>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Bennet Bo Fenner <53836821+bennetbo@users.noreply.github.com>
Before: (note the 1.3k in the upper right corner instead of 3 in the
second screenshot)
<img width="459" alt="Screenshot 2024-11-04 at 11 37 58 AM"
src="https://github.com/user-attachments/assets/64c06aff-f7d2-42a4-a767-0d7a4ba0f486">
Now:
<img width="631" alt="Screenshot 2024-11-04 at 11 38 11 AM"
src="https://github.com/user-attachments/assets/22af974d-915a-41e1-9ee0-f0622901e242">
This was intended to be a conservative estimate in case you pressed
Suggest Edits (and therefore might have an unpleasant surprise if you
were under the context limit but Suggest Edits failed with a "too much
context" error message anyway, after the Suggest Edits context got added
for you behind the scenes).
However, in retrospect this design created more [confusion in the common
case](https://github.com/zed-industries/zed/pull/19900#issuecomment-2453456569)
because it made it look like more context had been actually consumed
than what was actually consumed.
This does raise a potential design question for the future: the Suggest
Edits button adds more context at the last minute without ever
communicating that it's going to do that.
In the meantime it seems best to go back to the less-confusing way of
reporting the token counts, especially since only users of the
experimental flag could possibly press Suggest Edits anyway.
Release Notes:
- Fixed issue where initial token count was over-reported as 1.3k
instead of 3 (for the context string "You").
This a separate PR from https://github.com/zed-industries/zed/pull/19705
so we can revert it more easily if we want it back later.
Release Notes:
- Added "Suggest Edit" button to the assistant panel if
`"enable_experimental_live_diffs": true` is set in the `"assistant"`
section of `settings.json`. This button takes the place of the previous
`/workflow` command, but it is experimental and may change!
---------
Co-authored-by: Nathan Sobo <nathan@zed.dev>
Co-authored-by: Danilo Leal <67129314+danilo-leal@users.noreply.github.com>
I went with Cmd-Shift-Y on macOS (Ctrl-Shift-Y on Linux) for "yes accept
this individual hunk" - both are currently unused.
I went with Cmd-Shift-A on macOS (Ctrl-Alt-A on Linux) for "accept all
hunks" - both are unused. (Ctrl-Shift-A on Linux was taken, as is
Ctrl-Alt-Y, so although the pairing of Ctrl-Shift-Y and Ctrl-Alt-A isn't
necessarily obvious, the letters seem intuitive - "yes" and "all" - and
those key combinations don't conflict with anything.)
Release Notes:
- Added keybindings for applying hunks in Proposed Changes
<img width="247" alt="Screenshot 2024-10-25 at 12 47 00 PM"
src="https://github.com/user-attachments/assets/d6355621-ba80-4ee2-8918-b7239a4d29be">
Fixes an issue where evals were hitting "too many open files" errors
because we were adding (and detaching) new directory watches for each
project. Now we add those watches globally/at the worktree level, and we
store the tasks so they stop watching on drop.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
Co-authored-by: Piotr <piotr@zed.dev>
Co-authored-by: Piotr Osiewicz <24362066+osiewicz@users.noreply.github.com>
Now when an individual project eval fails, instead of panicking we add
it to a list of failures that we collect and report at the end (and make
the exit code nonzero).
Release Notes:
- N/A
Add `/auto` behind a feature flag that's disabled for now, even for
staff.
We've decided on a different design for context inference, but there are
parts of /auto that will be useful for that, so we want them in the code
base even if they're unused for now.
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
For future reference: WIP branch of copy/pasting a mixture of images and
text: https://github.com/zed-industries/zed/tree/copy-paste-images -
we'll come back to that one after landing this one.
Release Notes:
- You can now paste images into the Assistant Panel to include them as
context. Currently works only on Mac, and with Anthropic models. Future
support is planned for more models, operating systems, and image
clipboard operations.
---------
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Mikayla <mikayla@zed.dev>
Co-authored-by: Jason <jason@zed.dev>
Co-authored-by: Kyle <kylek@zed.dev>
We will soon need `semantic_index` to be able to use
`CompletionProvider`. This is currently impossible due to a cyclic crate
dependency, because `CompletionProvider` lives in the `assistant` crate,
which depends on `semantic_index`.
This PR breaks the dependency cycle by extracting two crates out of
`assistant`: `language_model` and `completion`.
Only one piece of logic changed: [this
code](https://github.com/zed-industries/zed/commit/922fcaf5a6076e56890373035b1065b13512546d#diff-3857b3707687a4d585f1200eec4c34a7a079eae8d303b4ce5b4fce46234ace9fR61-R69).
* As of https://github.com/zed-industries/zed/pull/13276, whenever we
ask a given completion provider for its available models, OpenAI
providers would go and ask the global assistant settings whether the
user had configured an `available_models` setting, and if so, return
that.
* This PR changes it so that instead of eagerly asking the assistant
settings for this info (the new crate must not depend on `assistant`, or
else the dependency cycle would be back), OpenAI completion providers
now store the user-configured settings as part of their struct, and
whenever the settings change, we update the provider.
In theory, this change should not change user-visible behavior...but
since it's the only change in this large PR that's more than just moving
code around, I'm mentioning it here in case there's an unexpected
regression in practice! (cc @amtoaer in case you'd like to try out this
branch and verify that the feature is still working the way you expect.)
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This implements the functionality (paired with @as-cii), but we weren't
sure what the clearest name would be for the action. It's essentially
the inverse of "quote selection" - but what's the opposite of quoting
the selection?
One idea:
* Rename "quote selection" to "Insert **into** assistant"
* Name this "Insert **from** assistant"
Release Notes:
- Added action to insert from assistant into editor (default keybinding:
`cmd-<` on macOS, `ctrl-<` on Linux)
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Bennet <bennet@zed.dev>
Release Notes:
- Now when an editor loses focus (e.g. from switching tabs) and then
gains focus again, it doesn't close the inline assist. Instead, it only
closes when you move the cursor outside of it, e.g. by clicking
somewhere else in its parent editor.
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Richard Feldman
