Files
hermes-agent/agent
Teknium e4b3b91b62 fix(compression): prune pre-checkpoint history on native compaction replay
Live verification (gpt-5.6 @ api.openai.com) proved the Responses server
renders NOTHING placed before a replayed compaction checkpoint: a fact
stated in a pre-checkpoint input item is invisible to the model, while the
same item after the checkpoint recalls perfectly. Hermes was replaying the
full pre-checkpoint transcript anyway — dead upload weight, and worse, every
plaintext user ask from before the boundary silently vanished from the
model's view, surviving only inside the opaque server summary. That is the
goal-drift failure mode reported against native compaction sessions.

Codex CLI never hits this because it rebuilds history client-side after
compaction, retaining user messages verbatim under a token budget. This
change is the wire-level equivalent: when a replayed checkpoint is present,
_chat_messages_to_responses_input restructures the input as

  [newest checkpoint run] + [retained pre-checkpoint user messages,
  newest-first within a 64K-token budget] + [post-checkpoint tail]

Histories without a checkpoint are returned unchanged, so non-native
sessions see a byte-identical wire.
2026-08-13 01:51:24 -07:00
..
2026-08-12 13:37:04 -04:00
2026-08-12 13:37:04 -04:00