mirror of
https://github.com/openai/codex.git
synced 2026-05-02 02:17:22 +00:00
## Summary - add the new `core/src/memories` module (phase-one parsing, rollout filtering, storage, selection, prompts) - add Askama-backed memory templates for stage-one input/system and consolidation prompts - add module tests for parsing, filtering, path bucketing, and summary maintenance ## Testing - just fmt - cargo test -p codex-core --lib memories::
1.2 KiB
1.2 KiB
Memory Consolidation
Consolidate Codex memories in this directory: {{ memory_root }}
Phase-1 inputs already prepared in this same directory:
raw_memories/contains per-thread raw memory markdown files.memory_summary.mdcontains a compact routing map from short summary -> thread id.
Consolidation goals:
- Read
memory_summary.mdfirst to route quickly, then open the most relevant files inraw_memories/. - Resolve conflicts explicitly:
- prefer newer guidance by default;
- if older guidance has stronger evidence, keep both with a verification note.
- Extract only reusable, high-signal knowledge:
- proven first steps;
- failure modes and pivots;
- concrete commands/paths/errors;
- verification and stop rules;
- unresolved follow-ups.
- Deduplicate aggressively and remove generic advice.
Expected outputs for this directory (create/update as needed):
MEMORY.md: merged durable memory registry for this CWD.skills/<skill-name>/...: optional skill folders when there is clear reusable procedure value.
Do not rewrite phase-1 artifacts except when adding explicit cross-references:
- keep
raw_memories/as phase-1 output; - keep
memory_summary.mdas the compact map generated from the latest summaries.