core: bundle settings diff updates into one dev/user envelope (#12417)

## Summary
- bundle contextual prompt injection into at most one developer message
plus one contextual user message in both:
  - per-turn settings updates
  - initial context insertion
- preserve `<model_switch>` across compaction by rebuilding it through
canonical initial-context injection, instead of relying on
strip/reattach hacks
- centralize contextual user fragment detection in one shared definition
table and reuse it for parsing/compaction logic
- keep `AGENTS.md` in its natural serialized format:
  - `# AGENTS.md instructions for {dirname}`
  - `<INSTRUCTIONS>...</INSTRUCTIONS>`
- simplify related tests/helpers and accept the expected snapshot/layout
updates from bundled multi-part messages

## Why
The goal is to converge toward a simpler, more intentional prompt shape
where contextual updates are consistently represented as one developer
envelope plus one contextual user envelope, while keeping parsing and
compaction behavior aligned with that representation.

## Notable details
- the temporary `SettingsUpdateEnvelope` wrapper was removed; these
paths now return `Vec<ResponseItem>` directly
- local/remote compaction no longer rely on model-switch strip/restore
helpers
- contextual user detection is now driven by shared fragment definitions
instead of ad hoc matcher assembly
- AGENTS/user instructions are still the same logical context; only the
synthetic `<user_instructions>` wrapper was replaced by the natural
AGENTS text format

## Testing
- `just fmt`
- `cargo test -p codex-app-server
codex_message_processor::tests::extract_conversation_summary_prefers_plain_user_messages
-- --exact`
- `cargo test -p codex-core
compact::tests::collect_user_messages_filters_session_prefix_entries
--lib -- --exact`
- `cargo test -p codex-core --test all
'suite::compact::snapshot_request_shape_pre_turn_compaction_strips_incoming_model_switch'
-- --exact`
- `cargo test -p codex-core --test all
'suite::compact_remote::snapshot_request_shape_remote_pre_turn_compaction_strips_incoming_model_switch'
-- --exact`
- `cargo test -p codex-core --test all
'suite::client::includes_apps_guidance_as_developer_message_when_enabled'
-- --exact`
- `cargo test -p codex-core --test all
'suite::client::includes_developer_instructions_message_in_request' --
--exact`
- `cargo test -p codex-core --test all
'suite::client::includes_user_instructions_message_in_request' --
--exact`
- `cargo test -p codex-core --test all
'suite::client::resume_includes_initial_messages_and_sends_prior_items'
-- --exact`
- `cargo test -p codex-core --test all
'suite::review::review_input_isolated_from_parent_history' -- --exact`
- `cargo test -p codex-exec --test all
'suite::resume::exec_resume_last_respects_cwd_filter_and_all_flag' --
--exact`
- `cargo test -p core_test_support
context_snapshot::tests::full_text_mode_preserves_unredacted_text --
--exact`

## Notes
- I also ran several targeted `compact`, `compact_remote`,
`prompt_caching`, `model_visible_layout`, and `event_mapping` tests
while iterating on prompt-shape changes.
- I have not claimed a clean full-workspace `cargo test` from this
environment because local sandbox/resource conditions have previously
produced unrelated failures in large workspace runs.
This commit is contained in:
Charley Cunningham
2026-02-26 00:12:08 -08:00
committed by GitHub
Unverified
parent 28bfbb8f2b
commit 07aefffb1f
47 changed files with 966 additions and 813 deletions
+13 -24
View File
@@ -1,21 +1,12 @@
use std::time::Duration;
use codex_protocol::models::ContentItem;
use codex_protocol::models::ResponseItem;
use crate::codex::TurnContext;
use crate::contextual_user_message::USER_SHELL_COMMAND_FRAGMENT;
use crate::exec::ExecToolCallOutput;
use crate::tools::format_exec_output_str;
pub const USER_SHELL_COMMAND_OPEN: &str = "<user_shell_command>";
pub const USER_SHELL_COMMAND_CLOSE: &str = "</user_shell_command>";
pub fn is_user_shell_command_text(text: &str) -> bool {
let trimmed = text.trim_start();
let lowered = trimmed.to_ascii_lowercase();
lowered.starts_with(USER_SHELL_COMMAND_OPEN)
}
fn format_duration_line(duration: Duration) -> String {
let duration_seconds = duration.as_secs_f64();
format!("Duration: {duration_seconds:.4} seconds")
@@ -48,7 +39,7 @@ pub fn format_user_shell_command_record(
turn_context: &TurnContext,
) -> String {
let body = format_user_shell_command_body(command, exec_output, turn_context);
format!("{USER_SHELL_COMMAND_OPEN}\n{body}\n{USER_SHELL_COMMAND_CLOSE}")
USER_SHELL_COMMAND_FRAGMENT.wrap(body)
}
pub fn user_shell_command_record_item(
@@ -56,15 +47,11 @@ pub fn user_shell_command_record_item(
exec_output: &ExecToolCallOutput,
turn_context: &TurnContext,
) -> ResponseItem {
ResponseItem::Message {
id: None,
role: "user".to_string(),
content: vec![ContentItem::InputText {
text: format_user_shell_command_record(command, exec_output, turn_context),
}],
end_turn: None,
phase: None,
}
USER_SHELL_COMMAND_FRAGMENT.into_message(format_user_shell_command_record(
command,
exec_output,
turn_context,
))
}
#[cfg(test)]
@@ -72,14 +59,16 @@ mod tests {
use super::*;
use crate::codex::make_session_and_context;
use crate::exec::StreamOutput;
use codex_protocol::models::ContentItem;
use pretty_assertions::assert_eq;
#[test]
fn detects_user_shell_command_text_variants() {
assert!(is_user_shell_command_text(
"<user_shell_command>\necho hi\n</user_shell_command>"
));
assert!(!is_user_shell_command_text("echo hi"));
assert!(
USER_SHELL_COMMAND_FRAGMENT
.matches_text("<user_shell_command>\necho hi\n</user_shell_command>")
);
assert!(!USER_SHELL_COMMAND_FRAGMENT.matches_text("echo hi"));
}
#[tokio::test]