agent: Do not send stale files notifications (#32974)

Removing it for two reasons:

1. We need a better implementation that doesn't hurt caching and doesn't
distracts the agent too much (see
https://github.com/zed-industries/zed/pull/32876 for more context)

2. Current insertion point of notifications doesn't play well with
Claude Thinking models (see
https://github.com/zed-industries/zed/issues/33000#issuecomment-2991709484)

I think we should get this code back in a form of a tool. But for now,
I'm dropping it to resolve recent issues.

Closes #33000

Release Notes:

- N/A
This commit is contained in:
Oleksiy Syvokon 2025-06-20 16:56:49 +03:00 committed by GitHub
parent c9e5ff21a1
commit b0f192ec2e
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 1 additions and 160 deletions

View file

@ -1,3 +0,0 @@
[The following is an auto-generated notification; do not reply]
These files have changed since the last read:

View file

@ -1,4 +1,3 @@
use std::fmt::Write as _;
use std::io::Write; use std::io::Write;
use std::ops::Range; use std::ops::Range;
use std::sync::Arc; use std::sync::Arc;
@ -37,6 +36,7 @@ use settings::Settings;
use thiserror::Error; use thiserror::Error;
use ui::Window; use ui::Window;
use util::{ResultExt as _, post_inc}; use util::{ResultExt as _, post_inc};
use uuid::Uuid; use uuid::Uuid;
use zed_llm_client::{CompletionIntent, CompletionRequestStatus}; use zed_llm_client::{CompletionIntent, CompletionRequestStatus};
@ -1389,8 +1389,6 @@ impl Thread {
request.messages[message_ix_to_cache].cache = true; request.messages[message_ix_to_cache].cache = true;
} }
self.attach_tracked_files_state(&mut request.messages, cx);
request.tools = available_tools; request.tools = available_tools;
request.mode = if model.supports_max_mode() { request.mode = if model.supports_max_mode() {
Some(self.completion_mode.into()) Some(self.completion_mode.into())
@ -1453,60 +1451,6 @@ impl Thread {
request request
} }
fn attach_tracked_files_state(
&self,
messages: &mut Vec<LanguageModelRequestMessage>,
cx: &App,
) {
let mut stale_files = String::new();
let action_log = self.action_log.read(cx);
for stale_file in action_log.stale_buffers(cx) {
if let Some(file) = stale_file.read(cx).file() {
writeln!(&mut stale_files, "- {}", file.path().display()).ok();
}
}
if stale_files.is_empty() {
return;
}
// NOTE: Changes to this prompt require a symmetric update in the LLM Worker
const STALE_FILES_HEADER: &str = include_str!("./prompts/stale_files_prompt_header.txt");
let content = MessageContent::Text(
format!("{STALE_FILES_HEADER}{stale_files}").replace("\r\n", "\n"),
);
// Insert our message before the last Assistant message.
// Inserting it to the tail distracts the agent too much
let insert_position = messages
.iter()
.enumerate()
.rfind(|(_, message)| message.role == Role::Assistant)
.map_or(messages.len(), |(i, _)| i);
let request_message = LanguageModelRequestMessage {
role: Role::User,
content: vec![content],
cache: false,
};
messages.insert(insert_position, request_message);
// It makes no sense to cache messages after this one because
// the cache is invalidated when this message is gone.
// Move the cache marker before this message.
let has_cached_messages_after = messages
.iter()
.skip(insert_position + 1)
.any(|message| message.cache);
if has_cached_messages_after {
messages[insert_position - 1].cache = true;
}
}
pub fn stream_completion( pub fn stream_completion(
&mut self, &mut self,
request: LanguageModelRequest, request: LanguageModelRequest,
@ -3229,106 +3173,6 @@ fn main() {{
); );
} }
#[gpui::test]
async fn test_stale_buffer_notification(cx: &mut TestAppContext) {
init_test_settings(cx);
let project = create_test_project(
cx,
json!({"code.rs": "fn main() {\n println!(\"Hello, world!\");\n}"}),
)
.await;
let (_workspace, _thread_store, thread, context_store, model) =
setup_test_environment(cx, project.clone()).await;
// Open buffer and add it to context
let buffer = add_file_to_context(&project, &context_store, "test/code.rs", cx)
.await
.unwrap();
let context =
context_store.read_with(cx, |store, _| store.context().next().cloned().unwrap());
let loaded_context = cx
.update(|cx| load_context(vec![context], &project, &None, cx))
.await;
// Insert user message with the buffer as context
thread.update(cx, |thread, cx| {
thread.insert_user_message("Explain this code", loaded_context, None, Vec::new(), cx)
});
// Create a request and check that it doesn't have a stale buffer warning yet
let initial_request = thread.update(cx, |thread, cx| {
thread.to_completion_request(model.clone(), CompletionIntent::UserPrompt, cx)
});
// Make sure we don't have a stale file warning yet
let has_stale_warning = initial_request.messages.iter().any(|msg| {
msg.string_contents()
.contains("These files changed since last read:")
});
assert!(
!has_stale_warning,
"Should not have stale buffer warning before buffer is modified"
);
// Modify the buffer
buffer.update(cx, |buffer, cx| {
// Find a position at the end of line 1
buffer.edit(
[(1..1, "\n println!(\"Added a new line\");\n")],
None,
cx,
);
});
// Insert another user message without context
thread.update(cx, |thread, cx| {
thread.insert_user_message(
"What does the code do now?",
ContextLoadResult::default(),
None,
Vec::new(),
cx,
)
});
// Create a new request and check for the stale buffer warning
let new_request = thread.update(cx, |thread, cx| {
thread.to_completion_request(model.clone(), CompletionIntent::UserPrompt, cx)
});
// We should have a stale file warning as the last message
let last_message = new_request
.messages
.last()
.expect("Request should have messages");
// The last message should be the stale buffer notification
assert_eq!(last_message.role, Role::User);
// Check the exact content of the message
let expected_content = "[The following is an auto-generated notification; do not reply]
These files have changed since the last read:
- code.rs
";
assert_eq!(
last_message.string_contents(),
expected_content,
"Last message should be exactly the stale buffer notification"
);
// The message before the notification should be cached
let index = new_request.messages.len() - 2;
let previous_message = new_request.messages.get(index).unwrap();
assert!(
previous_message.cache,
"Message before the stale buffer notification should be cached"
);
}
#[gpui::test] #[gpui::test]
async fn test_storing_profile_setting_per_thread(cx: &mut TestAppContext) { async fn test_storing_profile_setting_per_thread(cx: &mut TestAppContext) {
init_test_settings(cx); init_test_settings(cx);