The default-only guard corrected context_length to the real per-model cap
(e.g. 1M for claude-opus-4.7-1m) but left threshold_tokens pointing at the
ContextCompressor's stale value (computed from the global 232K cap → 197.2k
@ 85%). UI then showed 'auto-compress at 197.2k / 1M' which is misleading.
Rescale threshold_tokens by the real/orig ratio so the displayed trigger
reflects the actual window (e.g. ~850k @ 1M).
NOTE: this only corrects the SSE display payload. The real auto-compress
trigger lives inside ContextCompressor in hermes-agent (agent_init.py:1446
constructs it with the global cap). A full fix requires a parallel change
upstream — tracked separately.
_live_usage_snapshot() runs on every metering tick (~10x/sec while streaming).
The #3256 default-only guard recomputed get_model_context_length() there on
every tick for non-default models, which does a config read + potential
metadata/network probe — freezing claude-opus-4.7-1m streams while the default
model (4.8) stayed fast (guard not triggered for it). Resolve the real cap at
most once per stream via _real_ctx_cache. Backend-only, no frontend changes.
Re-applies cb0065eb + b34311b3 (context_length default-only guard) which were
dropped by the upgrade reset to v0.51.185. Fixes 4.7-1m context window showing
as stale global cap (232K) instead of real 1M metadata. Backend-only: touches
api/routes.py + api/streaming.py, zero frontend/render changes.
Instead of using len(visible_after)-1 (which points to the last visible
message and gets pushed behind the render window as more turns accumulate),
find the last [CONTEXT COMPACTION] marker in s.messages and compute the
anchor from visible messages before it.
This keeps the compression reference card at the correct boundary even
after 50+ subsequent turns have scrolled the render window past the old
anchor position.
Fixes a bug where the assistant's output message appeared to disappear
after automatic context compression because the reference card was placed
at the wrong position.
