# NousResearch/hermes-agent · Preserving model reasoning across turns, especially Kimi via OpenRouter

## Open now

- Nothing open on GitHub.

The members concern keeping model reasoning continuous across turns, reloads, and compaction. Two of them focus specifically on a Kimi reasoning-echo rule that lacks a model-name fallback when accessed through OpenRouter, including a proposed fix; the first is a broader statement about reasoning retention across providers.

## 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 #136358. Verify it addresses your case.
- **Recommended:** read PR #136358 (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: `unresolved`. 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evidence (`STRUCTURAL_SPANNING`): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.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

- #122303: kimi models via OpenRouter lose reasoning continuity: _REASONING_ECHO_RULES kimi rule has no model-name fallback

## Prior & competing PRs

- **#136358: Every model keeps its reasoning across turns, reloads and compaction on every provider that can read it**. merged to main · fix tier `verified`
  - maintainer direction: category `merged`
  - files: `agent/agent_init.py`, `agent/agent_runtime_helpers.py`, `agent/anthropic_adapter.py`, `agent/anthropic_endpoints.py`, `agent/anthropic_message_convert.py`, `agent/anthropic_thinking_policy.py`, `agent/anthropic_thinking_replay.py`, `agent/auxiliary_wire.py`, `agent/bedrock_adapter.py`, `agent/chat_completion_helpers.py`, `agent/chat_completion_helpers_relay.py`, `agent/chat_completion_helpers_tool_calls.py`, `agent/codex_responses_adapter.py`, `agent/context_compressor.py`, `agent/gemini_native_adapter.py`, `agent/message_metadata.py`, `agent/message_sanitization.py`, `agent/model_metadata.py`, `agent/reasoning_carriers.py`, `agent/reasoning_effort.py` (+35 more)
- **#122314: fix(agent): add model-substring fallback for kimi reasoning echo (#122303)**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `1f2b9e08644`, `#136358`, `#122303`

## Fix status

- Merged to main in PR #136358 (cited above). No release data is tracked. Verify downstream.
- Fix last verified (merged) `2026-10-11`.
- Subsystem: `agent`