[codex] add configurable token budget compaction reminder (#29255)

## Why

The token-budget feature reports coarse remaining-context milestones,
but it does not give the model a configurable wrap-up prompt before
automatic compaction. A strict threshold-crossing check can also miss
resumed or reconfigured windows that are already inside the threshold.

## What changed

- Add structured `[features.token_budget]` configuration for an absolute
`reminder_threshold_tokens` and bounded `reminder_message_template`;
`{n_remaining}` is expanded when the reminder is delivered.
- Compute remaining tokens against the next effective auto-compaction
boundary, including scoped `body_after_prefix` accounting and the full
context-window limit.
- Make reminder delivery level-triggered before and after sampling, with
one-shot state owned by `AutoCompactWindow` and re-armed on compaction,
`new_context`, restore, or history replacement.
- Leave the existing initial full-window token-budget context, 25/50/75%
notices, and token-budget tools unchanged.
- Persist the resolved feature configuration in the session config lock
and regenerate the config schema.

## Validation

- `just test -p codex-core token_budget`
- `just test -p codex-core
token_budget_reminder_emits_after_crossing_compaction_threshold`
- `just test -p codex-core auto_compact_window`
- `just test -p codex-core
lock_contains_prompts_and_materializes_features`
- `just test -p codex-features`
- `just test -p codex-config`
This commit is contained in:
pakrym-oai
2026-06-20 19:13:42 -07:00
committed by GitHub
Unverified
parent b6d6be2a84
commit 6df037d47f
15 changed files with 418 additions and 27 deletions
@@ -30,6 +30,7 @@ pub(super) struct AutoCompactWindow {
/// not the growth itself; server-observed usage replaces estimated
/// resume/recompute baselines when available.
prefill_input_tokens: Option<AutoCompactWindowPrefill>,
token_budget_reminder_delivered: bool,
}
impl AutoCompactWindow {
@@ -44,6 +45,7 @@ impl AutoCompactWindow {
},
new_context_window_requested: false,
prefill_input_tokens: None,
token_budget_reminder_delivered: false,
}
}
@@ -69,9 +71,14 @@ impl AutoCompactWindow {
self.ids.previous_window_id = Some(self.ids.window_id);
self.ids.window_id = Uuid::now_v7();
self.new_context_window_requested = false;
self.token_budget_reminder_delivered = false;
(self.window_number, self.ids)
}
pub(super) fn claim_token_budget_reminder(&mut self) -> bool {
!std::mem::replace(&mut self.token_budget_reminder_delivered, true)
}
pub(super) fn request_new_context_window(&mut self) {
self.new_context_window_requested = true;
}
@@ -154,6 +161,8 @@ mod tests {
);
assert_eq!(window.window_number(), 3);
assert_eq!(window.ids().window_id, restored_window_id);
assert!(window.claim_token_budget_reminder());
assert!(!window.claim_token_budget_reminder());
window.request_new_context_window();
assert!(window.take_new_context_window_request());
assert!(!window.take_new_context_window_request());
@@ -167,6 +176,7 @@ mod tests {
assert_eq!(ids.window_id.get_version_num(), 7);
assert_ne!(ids.window_id, restored_window_id);
assert!(!window.take_new_context_window_request());
assert!(window.claim_token_budget_reminder());
assert_eq!(
window.snapshot(),
+4
View File
@@ -148,6 +148,10 @@ impl SessionState {
self.auto_compact_window.snapshot()
}
pub(crate) fn claim_token_budget_reminder(&mut self) -> bool {
self.auto_compact_window.claim_token_budget_reminder()
}
pub(crate) fn auto_compact_window_number(&self) -> u64 {
self.auto_compact_window.window_number()
}