Watches a user-global `AGENTS.md` file alongside `settings.json` (at
`~/.config/zed/AGENTS.md` on macOS/Linux, `%APPDATA%\Zed\AGENTS.md` on
Windows) and includes its trimmed contents in the native agent's system
prompt.
This matches the pattern used by Codex (`CODEX_HOME/AGENTS.md`,
defaulting to `~/.codex/AGENTS.md`) and OpenCode
(`~/.config/opencode/AGENTS.md`): personal instructions live next to
other app config and apply across every project the user opens.
## Behavior
- Native Zed agent only. Not passed to ACP / external agents.
- Reads the local config dir, so SSH-remoted projects still get the
local user's personal `AGENTS.md` (project rules continue to come from
the remote workspace).
- Missing, empty, or whitespace-only files are silently treated as no
`AGENTS.md`.
- Read errors surface through the same notification UI as settings
errors, with a stable notification ID that's dismissed once the file
becomes readable again.
- The file is read in full, matching how existing project rules / repo
`AGENTS.md` files are loaded today.
## System prompt rendering
In the system prompt, the user-global `AGENTS.md` appears as `###
Personal AGENTS.md` immediately before `### Project Rules`, so the model
sees personal defaults first and project guidance later (project rules
take precedence on conflicts).
## Tests
- `user_agents_md` watcher: initial load, empty/whitespace ignored,
reacts to file edits.
- `SystemPromptTemplate`: renders personal `AGENTS.md` before project
rules; omits the section when no user `AGENTS.md` is present.
Closes AI-231
Release Notes:
- Added support for a global `AGENTS.md` file alongside `settings.json`
that is automatically included in the agent's instructions for every
project.
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Release Notes:
- N/A
We had a problem that became more evident to me in the newly-introduced
History tab in the Git panel where the avatars wouldn't show up for a
long time. Problem was that we were trying to render the avatar before
the GitHub user email came in, and that wouldn't work well because it
would 1) skip the fast CDN (which could get rate-limited fast), and 2)
cache the `None` result. So this improvement works by only attempting to
render the avatar when the email is available.
Release Notes:
- Git UI: Improve the display of user avatars in Git-related surfaces.
Follow-up to https://github.com/zed-industries/zed/pull/56500. Was
missing this ability very much :)
Release Notes:
- Git Panel: Added the ability to switch between the changes and history
tabs with the keyboard.
With long lines or a small viewport, the stage/restore buttons often
cover code.
Adds a setting to hide the buttons altogether. Defaults to showing them.
Release Notes:
- Added: Setting to hide Git Stage/Restore buttons
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes #ISSUE
Release Notes:
- N/A or Added/Fixed/Improved ...
The tool definition is very clearly contradicting the previous behavior.
Performance impact is unclear to me, we increase the work in a
potentially expensive loop, but it seems necessary to have both the
specified behavior from the tool definition, as well as the
heuristic/fallback for misbehaving models that seems to be intended.
Self-Review Checklist:
- [X] I've reviewed my own diff for quality, security, and reliability
- [X] Unsafe blocks (if any) have justifying comments
- [X] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [X] Tests cover the new/changed behavior
- [X] Performance impact has been considered and is acceptable
Closes#56225
Release Notes:
- Fixed tool paths preferring files in subdirectories named after the
project root
cc @SomeoneToIgnore
## Summary
Follow-up to #56409. I addressed the formatting issues and other stuff
from the previous PR
This keeps `editor.rs` around 15k lines.
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Release Notes:
- N/A
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
## Summary
Prevent undo grouping when an LSP completion includes extra edits so
that the completion and its extra edits are applied and reverted
atomically.
## Problem
When applying an LSP completion that also applies extra edits, the
editor may merge that completion into the surrounding undo group or
split the transaction while waiting for edits, causing undo to leave the
buffer in an inconsistent state.
## Solution
* Always block undo merging for completions
Closes #ISSUE
Release Notes:
- Prevent undo grouping when any LSP completion
A bit brute force, but it works.
<img width="1106" height="988" alt="image"
src="https://github.com/user-attachments/assets/d23f9a80-01c5-4ad3-a280-faf8b8bc9dbe"
/>
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Release Notes:
- N/A
---------
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Closes#55619
### Summary
- Route `buffer_search::UseSelectionForFind` through
`BufferSearchBar::deploy` instead of updating the query editor directly.
- Add an explicit seed-query override to `deploy`, so the Cmd-E action
can force `SeedQuerySetting::Always` while regular deploy callers
continue to pass `None` and respect the user’s
`seed_search_query_from_cursor` setting.
- By going through `deploy`, Cmd-E now also runs the search path that
keeps buffer-search navigation state in sync:
- shows/initializes the search bar for the active searchable item
- applies the seeded query via `search_suggested`
- calls `search`, which updates the query editor, search options, active
search query, search history, and macOS find pasteboard
- refreshes `searchable_items_with_matches` and `active_match_index`
- activates the current match after the search completes
- This ensures the subsequent Cmd-G action has the expected active
query, match list, search token, and active match index to select the
next result.
- Add a macOS-only end-to-end regression test using the default macOS
keymap with `simulate_keystrokes("cmd-e")` and
`simulate_keystrokes("cmd-g")`.
### Validation
- `cargo test -p search test_cmd_e_then_cmd_g_uses_selection_for_find`
- `cargo fmt --check --package search --package zed_actions`
- `./script/check-keymaps`
- `cargo check -p search`
- `cargo check -p workspace`
- `cargo check -p vim`
Release Notes:
- Fixed macOS Cmd-E/Cmd-G find behavior so Cmd-E seeds find from the
cursor or selection and Cmd-G advances through the newly seeded matches.
### Summary
Added a branch diff base branch picker that shows the selected diff base
and prioritizes branches on the active branch’s remote. This also
generalizes the branch picker’s checkout vs. select only behavior so its
UI/search/rendering can be reused by a future git graph picker, and
makes branch diff drop stale in progress base update tasks when the base
or repository changes for faster response.
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes #54050
Release Notes:
- git: Allow choosing branch diff base in branch diff view
---------
Co-authored-by: Remco Smits <djsmits12@gmail.com>
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [ ] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes#56677
Release Notes:
- Agent: Fix squashed message editor when `limit_content_width=false`
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Release Notes:
- Add the ability to create Terminal Threads in the Sidebar and Agent
Panel.
This PR adds a small refinement to the update button, removing
truncation from the version we display. Instead of just showing the
short version of a SHA, we now display it in full, providing more
detailed information that sometimes was lost. Also am making the button
disabled when it's in the checking, downloading, and installing states,
given clicking on it at these moments doesn't do anything.
Release Notes:
- N/A
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [ ] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes #ISSUE
Release Notes:
- Added the ability to input the "expected patch", i.e. what the model
should have predicted in the `edit prediction: rate completions` modal
Missed a test in #56194 with the same issue, and it bit me today :(
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes #ISSUE
Release Notes:
- N/A or Added/Fixed/Improved ...
This PR removes the seeding infrastructure from Collab.
We're already set up to just-in-time create users in local development
through Cloud.
Also updated the liveness probe for the health endpoint to use a
different query.
Closes CLO-763.
Release Notes:
- N/A
This reduces a rebuild time from editor (a legit one) from 28s to 18s.
For those of you who do use the debugger, there's a new `dbg` profile.
Self-Review Checklist:
- [ ] I've reviewed my own diff for quality, security, and reliability
- [ ] Unsafe blocks (if any) have justifying comments
- [ ] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [ ] Tests cover the new/changed behavior
- [ ] Performance impact has been considered and is acceptable
Closes #ISSUE
Release Notes:
- N/A or Added/Fixed/Improved ...
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [ ] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes #ISSUE
Release Notes:
- N/A or Added/Fixed/Improved ...
This PR removes the database access in the `Authorization` header
verification in Collab.
We already have the user returned from the call to Cloud, but were just
fetching the user from the database to get some additional fields.
We're now returning the additional fields we need from Cloud, so we can
just convert the user from the internal API response into a `User`
entity.
Closes CLO-762.
Release Notes:
- N/A
We currently run node on the JS wrapper around the native binaries
shipped with https://github.com/github/copilot-language-server-release.
According to their README, this is not required, and it seems like it is
just an option provided so that you can run the server with `npx`.
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [ ] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes#55891
Release Notes:
- Stopped relying on node for running the Copilot language server that
provides edit predictions. The system node version should no longer
affect whether Copilot edit predictions work in Zed
Clarifies that in the empty state, the "New Thread…" menu is exposed as
the agent selector button on the left, rather than the `+` button in the
top-right toolbar. Adds this note to the Getting Started sections for
Gemini CLI, Claude Agent, and Codex in `external-agents.md`, and
reorders the description in `agent-panel.md` to mention the empty state
first.
Release Notes:
- N/A
there was a "flicker" because it was showing "Terminal" for a
split-second
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Release Notes:
- N/A
#56472 broke Copilot chat
> Failed to connect to API: 400 Bad Request {"message":"cache_control:
Extra inputs are not permitted"}
This PR makes it so that we still use the legacy caching approach for
Copilot
Release Notes:
- N/A
Release Notes:
- Added support for opening a file with a line range selected in the
file finder using the `path:start-end` syntax (e.g. `file.rs:10-20`).
---------
Signed-off-by: Xiaobo Liu <cppcoffee@gmail.com>
Co-authored-by: Smit Barmase <heysmitbarmase@gmail.com>
- limit diagnostics sent to teacher
- improve parallelism in `ep` commands when using `--max-parallelism` by
only grouping by repo when instructed too (repo grouping is only useful
for context collection)
- update rejected and rated queries to also fetch settled editable
region
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes #ISSUE
Release Notes:
- N/A or Added/Fixed/Improved ...
Makes it so that "Restore checkpoint" shows up as soon as the model
performs an edit, instead of waiting until the turn ends
Release Notes:
- Fixed the agent panel to show "Restore checkpoint" as soon as the
agent starts editing instead of waiting until the turn ends
Closes:
- https://github.com/zed-industries/zed/discussions/56359
Self-Review Checklist:
- [ YES ] I've reviewed my own diff for quality, security, and
reliability
- [ N/A ] Unsafe blocks (if any) have justifying comments
- [ N/A ] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [ YES ] Tests cover the new/changed behavior
- [ YES ] Performance impact has been considered and is acceptable
Open to feedback or alternate approaches.
Could do `--ssh user@host:~code/proj` if preferred.
Or really cowboy it: regex replace `s|:~|/~|` and `s|:/|/|`; re-attempt
`Url::parse`
CC: @ConradIrwin
Release Notes:
- Add support for SCP style SSH urls: `zed ssh://user@host:~/code/proj`
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
------
This PR adds a settings panel above the CSV table with dropdown menus to
control rendering behavior, and a performance metrics overlay for
debugging.
Currently it's mostly used for me during dev phase, but before release
of CSV preview feature all dev-only options will be cleaned up, and
future features like "copy selected" will have their settings in this
bar
<img width="2260" height="1674" alt="image_2026-04-09_11-52-24"
src="https://github.com/user-attachments/assets/9039fd59-5c46-4be4-9f33-b9825a3cdce3"
/>
**What changed:**
- Adds `render_settings_panel()` method with dropdown menus for:
- **Rendering Mode**: Variable Height (multiline support) vs Uniform
Height (better performance)
- **Text Alignment**: Top vs Center vertical alignment within cells
- **Font Type**: UI Font vs Monospace for better readability (will be
exposed to user settings)
- **Experimental**: Popover menu with toggles for debug features
- Adds `render_performance_metrics_overlay()` method showing:
- CSV parsing duration
- Rendered row count and indices
- Positioned in bottom-right corner with semi-transparent styling
Context:
Will iterate on it with @Anthony-Eid
Release Notes:
- N/A
---------
Co-authored-by: Anthony Eid <anthony@zed.dev>
Treat `.agents/skills/` (project-local) and `~/.agents/skills/` (global)
as **sensitive paths**, on par with `.zed/` and the global config
directory. The agent's built-in editing tools (`edit_file`,
`write_file`, `create_directory`, `delete_path`, `move_path`,
`copy_path`) now require explicit user authorization before modifying
anything inside those paths, because the contents of skill files control
agent behavior.
This protection is worth landing on its own, ahead of Zed adding its own
skills support: other agents (e.g. Claude Code) already write skill
files into these locations, so a Zed installation may already have
skills on disk that should not be silently editable by the agent.
Also tightens the **pre-existing `.zed/` check** to compare path
components case-insensitively. macOS and Windows use case-insensitive
filesystems by default, so without this fix a malicious settings author
could bypass the local-settings classifier with `.ZED/settings.json`
(the canonicalized inode would match, but the path-component comparison
would miss it). The new `.agents/skills/` check has the same hazard and
now shares a single `component_matches_ignore_ascii_case` helper with
the `.zed/` check.
Introduces the `agent_skills` crate, scoped for now to just the path
constants and helpers (`global_skills_dir`,
`project_skills_relative_path`, `SKILL_FILE_NAME`) so the
tool-permission machinery can recognize the agent skills tree without
depending on a skill discovery / parsing / loading layer. Those will
land in follow-up PRs.
Closes AI-217
Release Notes:
- Agent: Require user confirmation before letting tools modify files
inside `.agents/skills/` (per-project) or `~/.agents/skills/` (global),
so skills installed by any agent are protected from unsolicited edits
---------
Co-authored-by: MartinYe1234 <52641447+MartinYe1234@users.noreply.github.com>
Co-authored-by: Martin Ye <martinye022@gmail.com>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
The Bedrock Converse API supports placing `CachePoint` blocks inside the
`system` field, but we were sending the system prompt as a single
`SystemContentBlock::Text`, which leaves the system tokens dependent on
whatever message-level breakpoint happens to fall within the 20-block
lookback window.
This widens `bedrock::Request.system` from `Option<String>` to
`Vec<BedrockSystemContentBlock>` and has `into_bedrock` emit
`[Text(system), CachePoint(Default)]` whenever the model supports prompt
caching. The system prompt now anchors its own cache prefix, on top of
the existing tool-list anchor and per-message breakpoint, so a stable
system prompt keeps producing cache hits even when earlier conversation
turns change.
Bedrock does not support automatic caching or the 1-hour TTL, so the
default 5-minute ephemeral cache is the only option for this provider.
Release Notes:
- Improved Bedrock prompt cache utilization by anchoring the system
prompt as its own cache prefix
Closes#55481
Adds Vim-mode access to the existing Helix jump-to-word overlay via `g
z`. We use `g z` because it is currently unassigned in Vim mode, while
`g w` is already used for rewrap.
Most of the implementation lives in `helix.rs` because the existing jump
overlay, label generation, and Helix/Vim modal behavior are currently
intertwined there. This keeps the change small and reuses the existing
navigation overlay logic instead of doing a broader refactor.
In Vim normal mode, jump labels behave like a cursor motion: selecting a
label moves the cursor to the start of the target word without selecting
it. In Vim visual mode, jump labels extend the selection like a Vim
word-start motion, preserving Vim’s inclusive visual-selection behavior.
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Release Notes:
- Added Vim-mode jump-to-word navigation on `g z`.
## Context
The process_tool_use closure used a single chars_read_so_far counter
shared across all REWRITE_SECTION_TOOL_NAME tool use events. When the
LLM produced multiple separate rewrite tool uses (different IDs), the
counter from the first tool use carried over into the second. If the
second tool use had a shorter replacement_text, the slice indexing
panicked with index out of bounds.
(This caused my first ever Zed panic today.)
Fix by tracking bytes-read per tool use ID using a HashMap keyed on
LanguageModelToolUseId. Also replace the direct slice index with .get()
so any remaining out-of-bounds case is handled gracefully instead of
panicking.
Looks like it hasn't been reported yet. At least I couldn't find any
related issues.
## How to Review
I hope the fix is pretty self-explanatory. We keep track of multiple
rewrite tools, which explains the extra ceremony around getting the
right character count, but apart from that it's pretty straightforward.
The performance impact should be neglible because:
1. Once the completion stream ends (or the task is dropped/cancelled),
the entire Arc<Mutex<HashMap>> is freed.
2. The number of entries is tiny. In a single completion request the LLM
will produce at most a handful of
REWRITE_SECTION_TOOL_NAME tool uses. Each entry is just an Arc<str> key
+ usize value.
So the HashMap grows for the duration of one handle_completion call and
is then dropped wholesale.
## Self-Review Checklist
<!-- Check before requesting review: -->
- [X] I've reviewed my own diff for quality, security, and reliability
- [X] Unsafe blocks (if any) have justifying comments
- [X] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [X] Tests cover the new/changed behavior
- [X] Performance impact has been considered and is acceptable
Release Notes:
- Fixed an out-of-bounds panic when the AI produced multiple inline
rewrites in a single completion.
Switches the Anthropic provider from hand-stamping `cache_control` onto
the last message content block over to Anthropic's top-level automatic
prompt caching, paired with explicit long-TTL (1h) anchors on the last
tool definition and on the system prompt.
The prefix order `tools` → `system` → `messages` satisfies Anthropic's
requirement that longer TTLs appear earlier in the prefix, so the static
prefix is cached for 1h (surviving idle gaps longer than the 5-minute
default) while the rapidly-changing conversation tail uses the
free-to-refresh 5-minute TTL via the top-level automatic breakpoint.
Three of the four available cache breakpoints are used (last tool,
system, automatic conversation), leaving one in reserve.
As a side benefit, this fixes a latent issue where the previous stamping
loop could place `cache_control` on a `Thinking` content block, which
the Anthropic API does not allow. Automatic caching is documented to
walk past ineligible blocks (including thinking) when selecting its
breakpoint, so we now delegate that responsibility to the server.
The new shape we send (when caching is enabled):
```json
{
"tools": [{ "...": "...", "cache_control": {"type": "ephemeral", "ttl": "1h"} }],
"system": [
{"type": "text", "text": "...", "cache_control": {"type": "ephemeral", "ttl": "1h"}}
],
"messages": [ /* no per-block cache_control */ ],
"cache_control": {"type": "ephemeral"}
}
```
Release Notes:
- Improved Anthropic prompt cache utilization, reducing latency and cost
for ongoing conversations
---------
Co-authored-by: Martin Ye <martinye022@gmail.com>
Resolves https://github.com/zed-industries/zed/issues/56518
This diff fixes code lens decorations continuing to paint into the
gutter when the editor is horizontally
scrolled. Code lens entries were rendered as flex editor blocks with
explicit gutter padding, which allowed
them to bypass the text viewport clipping used by content-only blocks.
With this change, code lens entries render as spacer-style custom blocks
instead. Spacer blocks already
scroll with buffer content and paint through the text-side mask, so the
code lens padding can be relative to
the editor text area and decorations clip at the left edge of the
viewport.
| Before | After |
| --- | --- |
| <video
src="https://github.com/user-attachments/assets/dc567ce5-ad04-4ba2-9156-ec4ec087bd41"
controls width="320"></video> | <video
src="https://github.com/user-attachments/assets/abb5ba2c-981f-4399-903b-0f8fa9a560e4"
controls width="320"></video> |
Release Notes:
- Fixed code lens decorations painting outside the editor viewport when
horizontally scrolling.
Code had been assuming (erroneously, but understandably) that
LlmApiToken::acquire would give them a valid token.
This is not true, as those tokens expire and you must call refresh
explicitly.
Add some helpers to do the retry for you, and rename acquire to cached
to be
clearer about the intent.
Closes #ISSUE
Release Notes:
- Fixed some rare cases where API requests would fail with Unauthorized
Follows similar approach as
https://github.com/zed-industries/zed/pull/55927
Adds a force-delete path to the worktree picker. Normal delete now
prompts when Git reports modified or untracked files, and
Alt/Option-delete can force delete directly.
Release Notes:
- Added support for force deleting worktrees that contain modified or
untracked files.
## Summary
Fixes scroll position inversion when content height changes during a
scrollbar drag (e.g. in the agent panel while streaming responses).
### Root cause
`set_offset_from_scrollbar` used the **live** `items.summary().height`
for `content_height` and computed a `drag_offset` correction (`live −
frozen`). However, `max_offset_for_scrollbar()` → `max_scroll_offset()`
already uses the **frozen** `scrollbar_drag_start_height` during drag.
This mismatch means:
1. The scrollbar computes thumb position using the frozen height range
2. `set_offset_from_scrollbar` converts that position using the live
height range
3. As content grows during drag, `drag_offset` increases
4. `(point.y - drag_offset).abs()` produces an incorrectly large value,
overshooting `scroll_max`
The `.abs()` call also masked the sign convention:
`compute_click_offset` (in `scrollbar.rs`) returns `-max_offset *
percentage` — a negative value — but `.abs()` converted it regardless of
sign, hiding the mismatch.
### Fix
- Use `scrollbar_drag_start_height` (frozen during drag) for
`content_height` in `set_offset_from_scrollbar`, matching
`max_scroll_offset()`. Both sides of the scrollbar mapping now use the
same height reference.
- Replace `(point.y - drag_offset).abs()` with `(-point.y)` to correctly
convert the negative scroll offset.
- Remove the symmetric `drag_offset` from
`scroll_px_offset_for_scrollbar`, which reported position back to the
scrollbar and suffered from the same inconsistency.
- Update existing tests to use negative offsets, matching the actual
values produced by the scrollbar component.
- Add a regression test that freezes height, grows content, drags the
scrollbar, and asserts the position is correct.
## Test plan
- [x] Existing tests updated to use correct sign convention (negative
offsets)
- [x] New regression test `test_scrollbar_drag_with_growing_content`:
verifies scroll position remains correct when content grows during a
scrollbar drag
- [ ] Manual: open agent panel, start a long generation, drag scrollbar
while content is streaming — position should track the thumb correctly
Release Notes:
- Fixed scrollbar position jumping or inverting when content height
changes during a scrollbar drag (e.g. in the agent panel while
streaming).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Smit Barmase <heysmitbarmase@gmail.com>
Adds streaming RPC forwarding to collab so guests can call
`GetInitialGraphData` and `SearchCommits` against a remote host project.
Previously these requests had no forwarder registered on the server and
would fail when invoked by a guest.
This mirrors the existing single-response forwarding pattern with new
analogues:
- `StreamResponse<R>` + `MessageContext::forward_request_stream`
- `Server::add_request_stream_handler`
- `forward_read_only_project_stream_request`, registered for both
messages
Also hardens both the unary and stream handlers to send
`respond_with_error` when a handler returns `Ok` without sending/ending
a response, so the client doesn't hang waiting for a reply that will
never arrive.
I added git graph collab integration tests for this as well.
Self-Review Checklist:
- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content is consistent with the [UI/UX
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable
Closes#55954
Release Notes:
- N/A
