Open SourceAutomated release watch

stagehand/server-v3 v3.7.4

Stagehand server-v3 v3.7.4 adds OpenAI Chat Completions support via openaiEndpointFormat, introduces evaluation trajectory grouping with Braintrust cross-linking, and updates documentation ownership rules.

Human read

Why this signal matters

Stagehand server-v3 v3.7.4 introduces support for OpenAI Chat Completions by exposing openaiEndpointFormat across both the core library and server-v3. Additionally, evaluation tooling has been updated to group trajectories by run and cross-link Braintrust experiments. Documentation ownership and review rules were also added.

Agent parse

Actionable summary

Browserbase Stagehand released stagehand/server-v3 v3.7.4, adding OpenAI Chat Completions compatibility via openaiEndpointFormat in core and server-v3, and grouping trajectories by run with Braintrust experiment cross-linking.

Agent usefulness
57/100
Confidence
96%
Canonical data
JSON + Markdown
Next actions

What builders should check

  • Verify compatibility or configure openaiEndpointFormat if utilizing OpenAI Chat Completions endpoints with Stagehand server-v3.
Classification

Tags and routing

browserbasebrowser-agentautomation
Related signals

Continue the thread