hermes-agent/website/docs
teknium1 400891d542
fix(caching): honor prompt_caching.enabled across model switch + fallback
@janrenz's PR #35862 added prompt_caching.enabled=false at init only. But
_anthropic_prompt_cache_policy re-derives _use_prompt_caching on every /model
switch (agent_runtime_helpers) and fallback-model swap (chat_completion_helpers),
which re-enabled markers and re-broke the strict proxy the toggle was meant to fix.

Move the kill switch into anthropic_prompt_cache_policy so it returns (False, False)
on every path. Drop the now-redundant init-time override (kept @janrenz's isinstance
hardening on the cache_ttl read). Add policy-level tests + docs for the toggle.

Follow-up to salvaged PR #35862.

(cherry picked from commit 36f9f50145)
2026-07-06 02:08:02 -07:00
..
developer-guide fix(caching): honor prompt_caching.enabled across model switch + fallback 2026-07-06 02:08:02 -07:00
getting-started Revert "Merge pull request #30179 from NousResearch/feat/iron-proxy" 2026-07-04 13:38:59 -07:00
guides feat(hooks): spill oversized hook-injected context to disk (#20468) 2026-07-05 13:51:26 -07:00
integrations feat(auth): make xAI Grok OAuth device-code-only, drop loopback login 2026-07-02 13:17:41 -07:00
reference feat(sessions): full filter surface for prune + bulk archive subcommand (#59327) 2026-07-05 22:04:52 -07:00
user-guide fix(caching): honor prompt_caching.enabled across model switch + fallback 2026-07-06 02:08:02 -07:00
index.mdx feat(docs): clarify platform support 2026-06-26 11:37:56 -07:00
user-stories.mdx docs(website): add User Stories and Use Cases collage page (#18282) 2026-04-30 23:56:59 -07:00