Repository Issue Activity (beta)

#7005 [Bug]: bifrost_active_requests drifts negative on streaming — per-chunk post-hooks decrement on a sticky StreamEndIndicator

#6971 [Bug]: retry scheduler ignores provider Retry-After / rate-limit headers it already captures and forwards

#6815 [Feature]: forward x-opencode-session on opencode requests

#7004 [Feature]: Extensible async video provider adapters for Seedance, Kling, MiniMax H3, and other APIs

#4916 [Bug]: Can't edit created mcps

#7003 [Bug]: Bedrock Converse assigns duplicate default name "document" to untitled document blocks → ValidationException

#6797 Docs still show pre-v1.5.0 virtual key budget/key fields as current

#6974 [Bug]: streaming first-chunk peek and drain wait ignore context, pinning workers for up to stream_idle_timeout

#6831 [Bug]: Responses-to-Chat mux drops non-streaming finish_reason

#6973 [Bug]: provider response headers leak across fallback boundaries (clearCtxForFallback misses ProviderResponseHeaders)

#6972 [Bug]: abandoned-request billing is a ~50% coin flip when a client disconnects mid-request

#6937 Bug: Dashboard and /api/logs display costs in raw micro-dollars instead of standard USD (e.g., displaying $1067 instead of $0.001067)

#6734 [Bug]: Periodic auto-reload on `/workspace/governance/customers` resets Customer edit form

#5876 [Bug]: key rotation is bounded by max_retries, so requests fail while healthy keys remain unused

#6964 [Bug]: retry backoff uses uninterruptible time.Sleep, holding worker slots for cancelled requests

#6967 [Bug]: shouldContinueWithFallbacks nil-derefs BifrostError.Error, crashing the process on a plugin-returned error

#6966 [Bug]: fallbacks silently re-target the primary provider for image edit / variation requests

#6624 Bedrock reasoning signature field is dropped for Anthropic models, which require it present

#6784 [Bug]: Chat completion stream hangs forever after finish_reason when upstream omits [DONE] but keeps sending heartbeats

#6646 [Feature]: Add native agent capability router plugin

#6955 [Bug]: `provider is shutting down` returned as HTTP 400 instead of a retryable 5xx

#6090 [Bug]: matview refresh gate is per-process, so every pod start re-runs a full refresh another replica just finished

#6914 [Bug]: Team current spend is reset after adjusting budget limit even when choosing Preserve Usage

#6933 [Bug]: Bedrock cachePoint markers are stripped when the model resolves to an application inference profile ID, silently disabling prompt caching

#6816 [Bug]: otel plugin exports gen_ai.input/output.messages in a non-semconv shape; Responses API instructions never reach the span