ioanbudea-OAI 979771476a Account for replay caches in encrypted compaction budgets (#53691) ...

## Why

Encrypted compaction payloads can include sampled-token replay caches that increase their size roughly threefold. Budgeting them as ordinary text can reject checkpoints that fit the model context budget.

## What changed

Use a shared estimate of one token per twelve ciphertext bytes, rounded up, for encrypted `Compaction` and `ContextCompaction` items in history and Guardian context budgeting. Guardian continues to count framing and metadata separately.

Check parent compactions against the configured token budget and a separate hard wire-size limit. Preserve the complete encrypted checkpoint when it fits. The token estimate remains a heuristic and can underestimate legacy checkpoints without replay caches.

## Testing

Add coverage for token-estimate rounding, parent compaction token and wire-size limits, and reviewer continuation with a cached compaction preserved intact.

GitOrigin-RevId: cc18ca29802491865167eef992830efcf93d97da

2026-10-13 15:05:16 +00:00

..

src

Account for replay caches in encrypted compaction budgets (#53691)

2026-10-13 15:05:16 +00:00

tests

Add opt-in output token replay for OpenAI requests (#52742)

2026-10-10 01:25:00 +00:00

BUILD.bazel

Add shared Guardian context primitives (#41392)

2026-08-28 19:02:26 +00:00

Cargo.toml

Keep Guardian transcript records structured through budget recovery (#51466)

2026-10-06 19:31:24 +00:00