Release Notes:

- Agent Beta: Fixed a delay when sending the first message in a new
thread

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.

This PR fixes the `assistant: open prompt library` action in the command
palette not opening the prompt library when the Assistant Panel did not
have focus.

Fixes https://github.com/zed-industries/zed/issues/28058.

Release Notes:

- assistant: Fixed `assistant: open prompt library` not opening the
prompt library when the Assistant Panel was not focused.

Closes #27223
Merges: #27996, #26734, #27949 

Release Notes:

- AWS Bedrock: Added advanced authentication strategies with:
  - Short lived credentials with Session Tokens 
  - AWS Named Profile
  - EC2 Identity, Pod Identity, Web Identity
- AWS Bedrock: Added Claude 3.7 Thinking support.
- AWS Bedrock: Adding Cross Region Inference for all combinations of
regions and model availability.
- Agent Beta: Added support for AWS Bedrock.

---------

Co-authored-by: Marshall Bowers <git@maxdeviant.com>

Cherry-picked copilot: Create Copilot directory if it does not exist
(#28157)

Closes https://github.com/zed-industries/zed/issues/27966

Issue:

- Copilot is failing to launch because the copilot directory is missing

<img width="497" alt="image"

src="https://github.com/user-attachments/assets/af35eb66-7e91-4dc6-a862-d1575da33b5b"
/>


<img width="943" alt="image"

src="https://github.com/user-attachments/assets/0b195c8c-52eb-42b9-bf36-40086398cc3f"
/>


Release Notes:

- copilot: Fixed an issue where GitHub Copilot would not install
properly if the directory was not present.

---------

Co-authored-by: Marshall Bowers <git@maxdeviant.com>

Co-authored-by: Richard Hao <richard@0xdev.dev>
Co-authored-by: Marshall Bowers <git@maxdeviant.com>

Cherry-picked agent: Fix opening configuration view from the model
selector (#28154)

This PR fixes an issue where opening the configuration view from the
model selector in the Agent (or inline assist) was not working properly.

Fixes https://github.com/zed-industries/zed/issues/28078.

Release Notes:

- Agent Beta: Fixed an issue where selecting "Configure" in the model
selector would not bring up the configuration view.

Co-authored-by: Marshall Bowers <git@maxdeviant.com>

Cherry-picked title_bar: Ensure git onboarding banner dismissal is
properly respected (#28147)

A user reported this issue [on

Discord](https://discord.com/channels/869392257814519848/873292398204170290/1357879959422636185).

The issue here only arises for users which recently installed Zed or had
previously not dismissed the Git Onboarding component. It was introduced
by https://github.com/zed-industries/zed/pull/27412, which made the
banner component reusable.

For every banner, there is a value stored in the KVP store when it was
first dismissed. For the git onboarding banner, this was
`zed_git_banner_dismissed_at` initially, but this key would have been
changed by the linked PR. A change would have resulted in the banner
being shown again for users who already dismissed the panel, so for the
special case of `Git Onboarding`, a check was added which ensured this
would not happen.

However, this check was only added for reading from the key from the DB
but not on writing the git onboarding dismissal it to the DB. Thus, if a
user who had not previously dismissed the panel opened Zed, we would
check for the old key to be present in the DB. Since that would not be
the case, the banner would be shown. If the user dismissed the panel, it
would be stored in the database with the new key. Thus, on a reopen of
Zed, the banner would again be shown since for the old key there would
still be no value present and users are unable to dismiss the panel.


This PR fixes this behavior by moving the check into the method that
generates the key. With this, users which were unaffected by the bug
will still not see the panel again. Users who would install Zed with
this change present will be able to properly dismiss the panel aswell.
Users which were affected by the bug need to dismiss the banner one more
time. That happens because I did not want to modify the dismissal check
to check for two keys (the original one and the new one), as it would
clutter the logic even more for this special case. If this would be
preferred, feel free to let me know.

Release Notes:

- Fixed an issue where dismissing the git onboarding banner would not be
persisted across sessions.

Co-authored-by: Finn Evers <dev@bahn.sh>

Tool use output wouldn't get rendered in some states.

Release Notes:

- N/A

This would cause the history view not to get refreshed immediately when
a thread was deleted

Release Notes:

- agent: Fixed a bug where the history view wouldn't refresh after
deleting a thread

#28102 introduced a bug where thread summaries wouldn't get generated
because they would get set to the default title instead of `None`.

Not adding a release note because the bug didn't make it to Preview.

Release Notes:

- N/A

A "Near" block acts similarly to a "Below" block, but can (if it's
height is <= one line height) be shown on the end of the preceding line
instead of adding an entire blank line to the editor.

You can test it out by pasting this into `go_to_diagnostic_impl` and
then press `F8`
```
        let buffer = self.buffer.read(cx).snapshot(cx);
        let selection = self.selections.newest_anchor();

        self.display_map.update(cx, |display_map, cx| {
            display_map.insert_blocks(
                [BlockProperties {
                    placement: BlockPlacement::Near(selection.start),
                    height: Some(1),
                    style: BlockStyle::Flex,
                    render: Arc::new(|_| {
                        div()
                            .w(px(100.))
                            .h(px(16.))
                            .bg(gpui::hsla(0., 0., 1., 0.5))
                            .into_any_element()
                    }),
                    priority: 0,
                }],
                cx,
            )
        });
        return;
```

Release Notes:

- N/A

---------

Co-authored-by: Antonio Scandurra <me@as-cii.com>

Previously, when comparing a block with an empty range to an empty query
range in non-inclusive mode, our binary search logic could end up
computing an inverted range, causing a panic.

This commit adds special casing when comparing empty blocks with empty
ranges.

cc @as-cii: I'm realizing that the approach to searching for the
intersecting replacement blocks makes some invalid assumptions about the
ordering of replace decorations. They aren't ordered at all by their end
range. @maxbrunsfeld and I are wondering if long term, we should remove
replace decorations and find another solution for folding buffers in
multi buffers.

Release Notes:

- Fixed an occasional panic that would occur when navigating to the next
change hunk with a pending inline transformation present.

Co-authored-by: Peter Tripp <petertripp@gmail.com>

This PR fixes a panic that could occur when opening the Agent diff from
the workspace (with the agent panel closed).

Release Notes:

- agent: Fixed a panic when running the `agent: open agent diff` command
with the Agent Panel closed.

Release Notes:

- agent: Tool call's input and output content are now rendered with
Markdown, which allows them to be selected and copied.

---------

Co-authored-by: Agus Zubiaga <hi@aguz.me>

Release Notes:

- agent: The scrollbar now automatically hides if there's no mouse
movement on the thread list.

---------

Co-authored-by: Agus Zubiaga <agus@zed.dev>
Co-authored-by: Agus Zubiaga <hi@aguz.me>

This PR adds tool calling support for GitHub Copilot Chat models.

Currently only supports the Claude family of models.

Release Notes:

- agent: Added tool calling support for Claude models in GitHub Copilot
Chat.

---------

Co-authored-by: Marshall Bowers <git@maxdeviant.com>

Using `shift-backspace` now because we need `backspace` for search

Release Notes:
- agent: Fix deleting threads in history via keyboard

Release Notes:

- agent: Disable tools that are redundant in the presence of the bash
tool

Cherry-picked Properly query remote ssh server for language servers by
name (#28124)

Follow-up of https://github.com/zed-industries/zed/pull/27775

Release Notes:

- N/A

Co-authored-by: Kirill Bulatov <kirill@zed.dev>

Cherry-picked jsx-tag-auto-close: Remove potential source of bugs and
panics (#28119)

Switch to using anchors for storing edited ranges rather than offsets as
they have to be used with multiple buffer snapshots

Release Notes:

- N/A

Co-authored-by: Ben Kunkle <ben@zed.dev>

Release Notes:

- Fixed a crash that could happen when expanding diff hunks with
multiple cursors in one hunk.

The description of the Batch Tool was still referring using `-` as a
seperator for tool names

Release Notes:

- N/A

Release Notes:

- agent: Remove `edit_files` tool  in favor of `find_replace`

Release Notes:

- agent: Add support for renaming threads

---------

Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>
Co-authored-by: Richard Feldman <oss@rtfeldman.com>

Release Notes:

- Enabled [token-efficient tool use
(beta)](https://docs.anthropic.com/en/docs/build-with-claude/tool-use/token-efficient-tool-use)
for Claude 3.7 Sonnet models

Closes: https://github.com/zed-industries/zed/issues/20582

Allows users to select a specific model for each AI-powered feature:
- Agent panel
- Inline assistant
- Thread summarization
- Commit message generation

If unspecified for a given feature, it will use the `default_model`
setting.

Release Notes:

- Added support for configuring a specific model for each AI-powered
feature

---------

Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>

Release Notes:

- N/A

---------

Co-authored-by: Michael Sloan <michael@zed.dev>

![CleanShot 2025-04-04 at 09 45
47@2x](https://github.com/user-attachments/assets/a8ec4086-f71e-4ff4-a5b3-4eb5d4c48294)


Release Notes:

- agent: Add search box to thread history

---------

Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>

Cherry-picked Clear path-based excerpt data properly (#28026)

Follow-up of https://github.com/zed-industries/zed/pull/27893

Release Notes:

- N/A

Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>

Co-authored-by: Kirill Bulatov <kirill@zed.dev>
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>

Cherry-picked Temporarily disable flaky conflicted-cherry-pick test
(#27950)

Closes #ISSUE

Release Notes:

- N/A

Co-authored-by: Cole Miller <cole@zed.dev>

Release Notes:

- Fixed a bug that would prevent rejecting certain agent edits.

This makes sure that we specify which lines the agent actually read,
avoids confusing scenarios such as:

<img width="642" alt="Screenshot 2025-04-04 at 10 22 10"
src="https://github.com/user-attachments/assets/2680c313-4f77-4971-8743-8e3f5327c18d"
/>

Here the agent starts out by actually only reading a certain amount of
lines when the first tool call happens, then it does a second tool call
to read the whole file. To the user this looks like to identical tool
calls.

Now:
<img width="621" alt="image"
src="https://github.com/user-attachments/assets/76222258-9cc8-4b7c-98c0-6d5cffb282f2"
/>
<img width="362" alt="image"
src="https://github.com/user-attachments/assets/293f2fc0-365d-4b84-8400-4c11474caeb8"
/>
<img width="420" alt="image"
src="https://github.com/user-attachments/assets/ca92493e-67ce-4d45-8f83-0168df575326"
/>



Release Notes:

- N/A

This ensures that we display @mentions and normal markdown links
differently:

<img width="670" alt="Screenshot 2025-04-04 at 11 07 51"
src="https://github.com/user-attachments/assets/0a4d0881-abb9-42a8-b3fa-912cd6873ae0"
/>


Release Notes:

- N/A

Release Notes:

- Added the ability to scroll to a file when clicking on it in the Agent
Panel review section.

This PR disables `parallel_tool_calls` for the models that support it,
as the Agent currently expects at most one tool use per turn.

It was a bit of trial and error to figure this out. OpenAI's API
annoyingly will return an error if passing `parallel_tool_calls` to a
model that doesn't support it.

Release Notes:

- N/A

This PR adds support for using tools to the OpenAI models.

Release Notes:

- agent: Added support for tool use with OpenAI models (Preview only).

This PR removes the unused `extract_tool_args_from_events` functions
that were defined in some of the LLM provider crates.

Release Notes:

- N/A

- Rename `TextField` -> `SingleLineInput`
- Add a component preview for `SingleLineInput`
- Apply `SingleLineInput` to the AddContextServerModal

Release Notes:

- N/A

---------

Co-authored-by: Agus Zubiaga <hi@aguz.me>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Danilo Leal <67129314+danilo-leal@users.noreply.github.com>

This makes context essentially work the same way as `read-file`,
increasing the likelihood of cache hits.

Just like with `read-file`, we'll notify the model when the user makes
an edit to one of the tracked files. In the future, we want to send a
diff instead of just a list of files, but that's an orthogonal change.


Release Notes:
- agent: Improved caching of files in context

---------

Co-authored-by: Antonio Scandurra <me@as-cii.com>

This PR adds the token count to the active thread view. It doesn't
behaves quite like Assistant 1 where it updates as you type, though; it
updates after you submit the message.

<img
src="https://github.com/user-attachments/assets/82d2a180-554a-43ee-b776-3743359b609b"
width="700" />

---

Release Notes:

- agent: Add token count in the thread view

---------

Co-authored-by: Agus Zubiaga <hi@aguz.me>

Release Notes:

- Fixed a rendering bug that caused context in the agent to not wrap
properly.

---------

Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
Co-authored-by: Zed AI <ai+claude-3.7@zed.dev>

### Context 

The Anthropic API fails if a request message contains a tool use and no
`Text` segments or it only contains empty `Text` segments. These are
cases that the model itself produces, but the API doesn't support
sending them back.

#27917 fixed this by appending "Using tool..." in the thread's message,
but this causes the actual conversation to include it, so it would
appear in the UI (we would actually display a gap because we never
rendered its markdown, but "Using tool..." would show up when the thread
was restored).

### Solution

We'll now only append this placeholder when we build the request, so the
API still sees it, but the UI/Thread doesn't.

Another issue we found is that the model starts mimicking these
placeholders in later tool uses which is undesirable. So unfortunately,
we had to add logic to filter them out.

Release Notes:

- agent: Improved rendering of tool uses without text

---------

Co-authored-by: Bennet <bennet@zed.dev>

This PR adds a button on the review tab empty state that toggles the
focus back to the agent panel so that users can keep iterating on the
thread that's active in the panel.

<img
src="https://github.com/user-attachments/assets/ace5cf93-8869-49bb-8106-e03a9e3c90f2"
width="700"/>

Release Notes:

- N/A

Previously the "Thinking..." step would show up as pending, even though
the user cancelled the generation:
<img width="672" alt="image"
src="https://github.com/user-attachments/assets/c9cdce0a-d827-4e23-96f5-b150465911a7"
/>


Release Notes:

- Fixed an issue where the thinking step would show up as pending even
when the generation was cancelled

This happened because of two reasons:

- `Workspace::recent_navigation_history` didn't include the current file
- The context picker added the current file to a exclude list

The latter was actually intentional because we already show the file in
the suggested context, but now that we actually have mentions, it's just
inconvenient not to have it there.

Release Notes:

- N/A

This was accidentally getting added due to increased feature
unification. We've manually excluded reqwest to go back to the desired
behavior: remote_server, doesn't depend on openssl.

Release Notes:

- N/A

Co-authored-by: Mikayla Maki <mikayla.c.maki@gmail.com>