Some notes:
- You can put the cursor on the start or end line with triple backticks,
it doesn't actually have to be inside the block.
- Placing the cursor outside of a code block does nothing.
- Code blocks are determined by counting triple backticks pairs from
either start or end of buffer, and nothing else.
- If you manually select something, the selection takes precedence over
any code blocks.
Release Notes:
- Added the ability to copy surrounding code blocks in the assistant
panel into the clipboard, or inserting them directly into the editor,
without manually selecting. Place cursor anywhere in a code block
(marked by triple backticks) and use the `assistant::CopyCode` action
(`cmd-k c` / `ctrl-k c`) to copy to the clipboard, or the
`assistant::InsertIntoEditor` action (`cmd-<` / `ctrl-<`) to insert into
editor.
---------
Co-authored-by: Thorsten Ball <mrnugget@gmail.com>
Co-authored-by: Bennet <bennet@zed.dev>
This PR adds a tool registry to hold tools that can be called by the
Assistant.
Currently we just have a `now` tool for retrieving the current datetime.
This is all behind the `assistant-tool-use` feature flag which currently
needs to be explicitly opted-in to in order for the LLM to see the
tools.
Release Notes:
- N/A
## Todo
* [x] Parse and present new XML output
* [x] Resolve new edits to buffers and anchor ranges
* [x] Surface resolution errors
* [x] Steps fail to resolve because language hasn't loaded yet
* [x] Treat empty `<symbol>` tag as None
* [x] duplicate assists when editing steps
* [x] step footer blocks can appear *below* the following message header
block
## Release Notes:
- N/A
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
Co-authored-by: Peter <peter@zed.dev>
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
This PR fixes two issues:
1. The prompt library window didn't set an `app_id` on Linux, which
caused it to be missing the Zed logo
2. A dangling reference to the window in the Wayland client code, which
caused the prompt library window not to close. See:
https://github.com/zed-industries/zed/pull/13201
Release Notes:
- Linux: Fixed the prompt library not closing on Wayland
---------
Co-authored-by: Junkui Zhang <364772080@qq.com>
Fixes:
- [x] an issue where directories would only match by prefix, causing
both a directory and a file to be matched if in the same directory
- [x] An issue where you could not continue a file completion when
selecting a directory, as `tab` on a file would always run the command.
This effectively disabled directory sub queries.
- [x] Inconsistent rendering of files and directories in the slash
command
Release Notes:
- N/A
---------
Co-authored-by: max <max@zed.dev>
This commit proposes the addition of "context serveres" and the
underlying protocol (model context protocol). Context servers allow
simple definition of slash commands in another language and running
local on the user machines. This aims to quickly prototype new commands,
and provide a way to add personal (or company wide) customizations to
the assistant panel, without having to maintain an extension. We can
use this to reuse our existing codebase, with authenticators, etc and
easily have it provide context into the assistant panel.
As such it occupies a different design space as extensions, which I
think are
more aimed towards long-term, well maintained pieces of code that can be
easily distributed.
It's implemented as a central crate for easy reusability across the
codebase
and to easily hook into the assistant panel at all points.
Design wise there are a few pieces:
1. client.rs: A simple JSON-RPC client talking over stdio to a spawned
server. This is
very close to how LSP work and likely there could be a combined client
down the line.
2. types.rs: Serialization and deserialization client for the underlying
model context protocol.
3. protocol.rs: Handling the session between client and server.
4. manager.rs: Manages settings and adding and deleting servers from a
central pool.
A server can be defined in the settings.json as:
```
"context_servers": [
{"id": "test", "executable": "python", "args": ["-m", "context_server"]
]
```
## Quick Example
A quick example of how a theoretical backend site can look like. With
roughly 100 lines
of code (nicely generated by Claude) and a bit of decorator magic (200
lines in total), one
can come up with a framework that makes it as easy as:
```python
@context_server.slash_command(name="rot13", description="Perform a rot13 transformation")
@context_server.argument(name="input", type=str, help="String to rot13")
async def rot13(input: str) -> str:
return ''.join(chr((ord(c) - 97 + 13) % 26 + 97) if c.isalpha() else c for c in echo.lower())
```
to define a new slash_command.
## Todo:
- Allow context servers to be defined in workspace settings.
- Allow passing env variables to context_servers
Release Notes:
- N/A
---------
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>
I need this to refine our prompts on the fly as I work.
Release Notes:
- Templates for prompts driving inline transformation in editors and the
terminal can now be overridden in the `~/.config/zed/prompts/templates`
directory. This is an advanced feature, and prevents you from getting
upstream changes. It's intended for use by Zed developers.
Release Notes:
- Improved streaming git diffs to be less jumpy during inline assistant
generation
---------
Co-authored-by: Antonio Scandurra <antonio@zed.dev>
This PR makes it so we normalize the line endings for prompts to LFs
(`\n`) when we load a prompt from the library.
In some cases, prompts could end up with CRLF (`\r\n`) line endings.
When these prompts were used with the `/prompt` slash command and
summarily run, the prompt text would be converted into a rope, causing
the line endings to be normalized to LFs.
However, this would happen _after_ the ranges for the
`SlashCommandOutputSection`s were computed based on the text that still
contained the CRLFs. This would then cause these ranges to be invalid
for the text with the normalized endings, resulting in a panic when
converting them to anchors.
Fixes https://github.com/zed-industries/zed/issues/15652.
Release Notes:
- N/A
Co-authored-by: Max <max@zed.dev>
- [x] bug: setting a key doesn't update anything
- [x] show high-level text on configuration page to explain what it is
- [x] show "everything okay!" status when credentials are set
- [x] maybe: add "verify" button to check credentials
- [x] open configuration page when opening panel for first time and
nothing is configured
- [x] BUG: need to fix empty assistant panel if provider is `zed.dev`
but not logged in
Co-Authored-By: Thorsten <thorsten@zed.dev>
Release Notes:
- N/A
---------
Co-authored-by: Thorsten <thorsten@zed.dev>
Co-authored-by: Nate Butler <iamnbutler@gmail.com>
Co-authored-by: Thorsten Ball <mrnugget@gmail.com>
This UI change is behind a `ZedPro` feature flag so that it won't be
visible until we're ready to launch that service.
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
Co-authored-by: Marshall <marshall@zed.dev>
We also eliminate the `completion` crate and moved its logic into
`LanguageModelRegistry`.
Release Notes:
- N/A
---------
Co-authored-by: Nathan <nathan@zed.dev>
This adds the optional `PRESERVED_KEYS` constant to the `Settings`
trait,
which allows users of the trait to specify which keys should be written
to
the settings file, even if their current value matches the default
value.
That's useful for tagged settings that have, for example, a `"version"`
field
that should always be present in the user settings file, so we can then
reparse
the user settings based on the version.
Co-Authored-By: Thorsten <thorsten@zed.dev>
Release Notes:
- N/A
---------
Co-authored-by: Thorsten <thorsten@zed.dev>
This PR removes the feature flag for the terminal inline assistant,
making it available to everyone.
Release Notes:
- Use Ctrl+Enter to summon the inline assistant in the terminal, which
let's you generate terminal commands based on your description
([demo](https://twitter.com/thorstenball/status/1814241447383605329))
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](922fcaf5a6 (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>
Note that this shouldn't have any visible user-facing behavior yet. The
feature is incomplete but we wanna merge early to avoid a long-running
branch.
Release Notes:
- N/A
---------
Co-authored-by: Nathan <nathan@zed.dev>
This pull request introduces collaboration for the assistant panel by
turning `Context` into a CRDT. `ContextStore` is responsible for sending
and applying operations, as well as synchronizing missed changes while
the connection was lost.
Contexts are shared on a per-project basis, and only the host can share
them for now. Shared contexts can be accessed via the `History` tab in
the assistant panel.
<img width="1819" alt="image"
src="https://github.com/zed-industries/zed/assets/482957/c7ae46d2-cde3-4b03-b74a-6e9b1555c154">
Please note that this doesn't implement following yet, which is
scheduled for a subsequent pull request.
Release Notes:
- N/A
This PR begins the process of making the backing infrastructure for the
`/rustdoc` command more generic such that it can be applied to
additional documentation providers.
In this PR we:
- Rename the `rustdoc` crate to `indexed_docs` as a more general-purpose
name
- Start moving rustdoc-specific functionality into
`indexed_docs::providers::rustdoc`
- Add an `IndexedDocsRegistry` to hold multiple `IndexedDocsStore`s (one
per provider)
We haven't yet removed the rustdoc-specific bits in the `DocsIndexer`.
That will follow soon.
Release Notes:
- N/A
This PR extracts the definition of the various Zed paths out of `util`
and into a new `paths` crate.
`util` is for generic utils, while these paths are Zed-specific. For
instance, `gpui` depends on `util`, and it shouldn't have knowledge of
these paths, since they are only used by Zed.
Release Notes:
- N/A
Closes#4424.
A few design decisions that may need some rethinking or later PRs:
* Other providers have a check for authentication. I use this
opportunity to fetch the models which doubles as a way of finding out if
the Ollama server is running.
* Ollama has _no_ API for getting the max tokens per model
* Ollama has _no_ API for getting the current token count
https://github.com/ollama/ollama/issues/1716
* Ollama does allow setting the `num_ctx` so I've defaulted this to
4096. It can be overridden in settings.
* Ollama models will be "slow" to start inference because they're
loading the model into memory. It's faster after that. There's no UI
affordance to show that the model is being loaded.
Release Notes:
- Added an Ollama Provider for the assistant. If you have
[Ollama](https://ollama.com/) running locally on your machine, you can
enable it in your settings under:
```jsonc
"assistant": {
"version": "1",
"provider": {
"name": "ollama",
// Recommended setting to allow for model startup
"low_speed_timeout_in_seconds": 30,
}
}
```
Chat like usual
<img width="1840" alt="image"
src="https://github.com/zed-industries/zed/assets/836375/4e0af266-4c4f-4d9e-9d74-1a91f76a12fe">
Interact with any model from the [Ollama
Library](https://ollama.com/library)
<img width="587" alt="image"
src="https://github.com/zed-industries/zed/assets/836375/87433ac6-bf87-4a99-89e1-96a93bf8de8a">
Open up the terminal to download new models via `ollama pull`:
This PR overhauls the HTML to Markdown conversion functionality in order
to make it more pluggable. This will ultimately allow for supporting a
variety of different HTML input structures (both natively and via
extensions).
As part of this, the `rustdoc_to_markdown` crate has been renamed to
`html_to_markdown`.
The `MarkdownWriter` now accepts a list of trait objects that can be
used to drive the conversion of the HTML into Markdown. Right now we
have some generic handler implementations for going from plain HTML
elements to their Markdown equivalents, as well as some rustdoc-specific
ones.
Release Notes:
- N/A
This could still use some improvement UI-wise but the user experience
should be a lot better.
- [x] Show in "Window" application menu
- [x] Load prompt as it's selected in the picker
- [x] Refocus picker on `esc`
- [x] When creating a new prompt, if a new prompt already exists and is
unedited, activate it instead
- [x] Add `/default` command
- [x] Evaluate /commands on prompt insertion
- [x] Autocomplete /commands (but don't evaluate) during prompt editing
- [x] Show token count using the settings model, right-aligned in the
editor
- [x] Picker
- [x] Sorted alpha
- [x] 2 sublists
- Default
- Empty state: Star a prompt to add it to your default prompt
- Otherwise show prompts with star on hover
- All
- Move prompts with star on hover
Release Notes:
- N/A
Using the file system as a database seems like it's easy, but it's
actually a real pain. I'd like to use LMDB to store the prompts locally
so we have more control. We can always add an export option, but I want
the source of truth to be somewhere other than the file system.
So far, I have a PromptStore which is global to the application and can
be initialized on startup. Then there's a `PromptLibrary` which is
intended to be the root of a new kind of Zed window. I haven't actually
seen pixels yet, but I've sketched out the basics needed to create a new
prompt, save, etc.
Still lots to figure out but the foundations of being backed by a DB and
rendering in an independent window are in place.
/cc @iamnbutler @as-cii
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
This is still pretty raw, so I'd like to hold off on shipping it to all
users.
Release Notes:
- Hide the prompt library for non-staff until it is in a more complete
state.
This PR adds a `/rustdoc` slash command for retrieving and inserting
rustdoc docs into the Assistant.
Right now the command accepts the crate name as an argument and will
return the top-level docs from `docs.rs`.
Release Notes:
- N/A
This PR extracts the `SlashCommand` trait (along with the
`SlashCommandRegistry`) from the `assistant` crate.
This will allow us to register slash commands from extensions without
having to make `extension` depend on `assistant`.
Release Notes:
- N/A
Restructure prompts & the prompt library.
- Prompts are now written in markdown
- The prompt manager has a picker and editable prompts
- Saving isn't wired up yet
- This also removes the "Insert active prompt" button as this concept doesn't exist anymore, and will be replaced with slash commands.
I didn't staff flag this, but if you do play around with it expect it to still be pretty rough.
Release Notes:
- N/A
---------
Co-authored-by: Nathan Sobo <1789+nathansobo@users.noreply.github.com>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Tasks
* [x] remove old flaps and output when editing a slash command
* [x] the completing a command name that takes args, insert a space to
prepare for typing an arg
* [x] always trigger completions when typing in a slash command
* [x] don't show line numbers
* [x] implement `prompt` command
* [x] `current-file` command
* [x] state gets corrupted on `duplicate line up` on a slash command
* [x] exclude slash command source from completion request
Next steps:
* show output token count in flap trailer
* add `/project` command that matches project ambient context
* delete ambient context
Release Notes:
- N/A
---------
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
This uses Jaro-Winkler similarity for now, which seemed to produce
pretty good results in my tests. We can easily swap it with something
else if needed.
Release Notes:
- N/A
### Todo
* [x] tuck the new system prompt away somehow
* for now, we're treating it as built-in, and not editable. once we have
a way to fold away default prompts, let's make it a default prompt.
* [x] when applying edits, re-parse the edit from the latest content of
the assistant buffer (to allow for manual editing of edits)
* [x] automatically adjust the indentation of edits suggested by the
assistant
* [x] fix edit row highlights persisting even when assistant messages
with edits are deleted
* ~adjust the fuzzy search to allow for small errors in the old text,
using some string similarity routine~
We decided to defer the fuzzy searching thing to a separate PR, since
it's a little bit involved, and the current functionality works well
enough to be worth landing. A couple of notes on the fuzzy searching:
* sometimes the assistant accidentally omits line breaks from the text
that it wants to replace
* when the old text has hallucinations, the new text often contains the
same hallucinations. so we'll probably need to use a more fine-grained
editing strategy where we perform a character-wise diff of the old and
new text as reported by the assistant, and then adjust that diff so that
it can be applied to the actual buffer text
Release Notes:
- Added the ability to request edits to project files using the
assistant panel.
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Nathan <nathan@zed.dev>
This PR adds the beginnings of current project context to the Assistant.
Currently it supports reading a `Cargo.toml` file and using that to get
some basic information about the project, and its dependencies:
<img width="1264" alt="Screenshot 2024-05-14 at 6 17 03 PM"
src="https://github.com/zed-industries/zed/assets/1486634/cc8ed5ad-0ccb-45da-9c07-c96af84a14e3">
Release Notes:
- N/A
---------
Co-authored-by: Nate <nate@zed.dev>
Release Notes:
- Added support for interacting with Claude in the assistant panel. You
can enable it by adding the following to your `settings.json`:
```json
"assistant": {
"version": "1",
"provider": {
"name": "anthropic"
}
}
```
