 Joel Hooks
|
d6094e4b89
|
ADR-0203: Compaction recovery pipeline
Three-stage approach:
1. Continuous extraction — turn_end hook monitors context %, starts
extracting decisions/failures/preferences at 40% (warm), flushes
to Redis + memory at 60% (hot). Zero latency on compaction itself.
2. Task checkpoint — structured Redis doc with currentTask, files,
decisions, pre-computed recall queries. Updated incrementally.
3. Post-compaction pointer injection — ~100 token signpost message
with task, files, decisions, and recall query refs. Pointers not
content — preserves context budget (40% already consumed post-compact).
Key insight: post-compaction context is 36-55% full. Every token of
injection competes with actual work. Pointers + on-demand recall beats
dumping observation text.
|
2026-03-04 08:46:20 -08:00 |
|