# NousResearch/hermes-agent · Stale reasoning leakage across turns or models

## Open now

- Nothing open on GitHub.

These items address a bug where reasoning items or encrypted content from one model or turn incorrectly persist and appear in subsequent turns or for co-hosted models on OpenAI-compatible endpoints. The work involves scoping reasoning replay to the originating model and handling foreign message IDs.

## TL;DR

- 3 member(s): 0 open issue(s), 1 closed issue(s), 0 open PR(s), 1 merged PR(s), 1 PR(s) closed without merging.
- A fix was merged to main in PR #111181. Verify it addresses your case.
- **Recommended:** read PR #111181 (and any competing PRs below) before filing. Confirm the merged change covers your symptom.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `resolved` · deciding PR(s): `pr-111181`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_SPANNING`): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.
- Pairwise link density `0.67`: the share of member pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. `1.00` means every pair is linked.

## Closed / duplicate issues

- #95834: \[Bug\]: Stale encrypted_content reasoning items leak across turns on OpenAI-compatible endpoints using codex_responses api_mode

## Prior & competing PRs

- **#111181: Codex Responses replay drops foreign message ids and scopes reasoning replay to the issuing model (#92311, #95834, salvage #92325 #95849)**. merged to main · fix tier `verified`
  - maintainer direction: category `merged`
  - files: `agent/auxiliary_client.py`, `agent/codex_responses_adapter.py`, `agent/error_classifier.py`, `agent/transports/codex.py`, `tests/agent/test_codex_responses_adapter.py`, `tests/agent/test_error_classifier.py`, `tests/agent/transports/test_codex_transport.py`
- **#95849: fix(agent): stop co-hosted models from inheriting stale reasoning**. closed without merge
  - maintainer direction: category `merged_via`; implication `superseded`; refs `14a93d67eb`, `#111181`

## Fix status

- Merged to main in PR #111181 (cited above). No release data is tracked. Verify downstream.
- Fix last verified (merged) `2026-09-15`.