# NousResearch/hermes-agent · LiteLLM prompt caching fixes for OpenAI-compatible wire

## Open now

- Nothing open on GitHub.

These issues and pull requests address a bug where prompt caching was not engaging for LiteLLM when using the OpenAI-compatible wire protocol, with fixes targeting both Claude and OpenAI-compatible paths.

## TL;DR

- 6 member(s): 0 open issue(s), 1 closed issue(s), 0 open PR(s), 1 merged PR(s), 4 PR(s) closed without merging.
- A fix was merged to main in PR #87517. Verify it addresses your case.
- **Recommended:** read PR #87517 (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-87517`. 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evidence (`STRUCTURAL_SPANNING`): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.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 `0.33`: 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

- #84506: \[Bug\]: Prompt caching never engages for LiteLLM on the OpenAI-compatible wire

## Prior & competing PRs

- **#87517: fix(caching): engage prompt caching for LiteLLM Claude on the OpenAI wire (envelope layout)**. merged to main · fix tier `verified`
  - maintainer direction: category `merged`; implication `superseded`
  - files: `agent/agent_runtime_helpers.py`, `tests/run_agent/test_anthropic_prompt_cache_policy.py`
- **#84544: fix(agent): grant prompt cache for Claude on LiteLLM OpenAI wire**. closed without merge
  - maintainer direction: category `duplicate_of`; implication `duplicate`; refs `#84550`
- **#84550: fix(agent): enable prompt caching for LiteLLM OpenAI-compatible wire (#84506)**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#87039`, `#84550`, `#84982`, `#87039`, `#84506`
- **#84982: fix(caching): engage prompt caching for LiteLLM Claude on the OpenAI wire**. closed without merge
  - maintainer direction: category `merged_via`; implication `superseded`; refs `ff4df5e54d`, `#87517`
- **#87039: fix(caching): combined LiteLLM OpenAI-wire prompt caching from #84550 and #84982 (#84506)**. closed without merge
  - maintainer direction: category `implemented_elsewhere`; implication `superseded`; refs `#87517`, `#84982`, `#77217`, `#69512`, `#87517`, `#87517`, `#87517`

## Fix status

- Merged to main in PR #87517 (cited above). No release data is tracked. Verify downstream.
- Fix last verified (merged) `2026-08-16`.
- Subsystem: `agent`