Analytics > Topics: 'Threads Per Topic' and 'Most Discussed Topics' label series with raw topic ids instead of topic names
Problèmes signalés
Dashboards: batch a board's widget queries with tRPC httpBatchStreamLink
Suggestions
Bulk imports on small self-hosted installs: eval-trigger retries dominate ClickHouse load
Experiment workbench shows stale saved-dataset rows after the dataset changes elsewhere (e.g. version restore)
Instant Evals explorer says "Nothing matches" when every row was skipped
Feature request: LangWatch Evaluation Nodes not visible in n8n Cloud
Langy shows the same vague 'model provider rejected that' for not-deployed and unsupported-parameter errors
Langy sends reasoning.effort to non-reasoning OpenAI models (gpt-4.1 → 400)
Model picker offers the full OpenAI catalog when the OpenAI provider has a custom base URL
First trace of a new project can get no trace summary
pi: one trace per process, so a multi-prompt session shows one turn
Feature request: Improve resiliency at trace endpoint
[BUG] App fails to load when self-hosted in a subpath with a reverse proxy
Dashboards: per-dashboard access, and concurrent widget queries on small installs
Remove the "Checked as" line from the scenario run drawer
LLM Judge Quality Analysis: Measure the Reliability of Your LLM Evaluators
Scenario setup and teardown hooks for HTTP agents
Coding agent traces: say when message content was not captured
Coding agent traces are named after their root span
Trace renderings sent to judges: six defects (span typing, bounded digest, system prompt, RAG contexts, relative time, legacy digest)
fix(instant-evals): own decision models run without the hosted flag, and Evals Preview hides where it can't run
fix(analytics): read the stock trace widgets from trace_metrics
fix(traces): count the span type facet only when its section opens
feat: agent and evaluator versioning with deploy labels, plus experiment and dependency fixes
feat(coding-agent): capture Grok CLI sessions through its native OpenTelemetry
feat(aigateway): serve OpenAI Decisions API on /v1/decisions
fix(sso): self-hosted sign-in sends colleagues straight to the sole live connection